View in EDS HTML Full Text PDF Full Text

Unraveling Coherence: The Impact of Image-Relatedness in Concept Learning

Saved in:
Bibliographic Details
Title: Unraveling Coherence: The Impact of Image-Relatedness in Concept Learning
Language: English
Authors: Tristan H. S. de Jonge (ORCID 0009-0003-9983-4223), Anna Berti (ORCID 0009-0007-2837-6770), Sanne van Schijndel, Margot van Wermeskerken (ORCID 0000-0003-3883-9875), Ellen Kok (ORCID 0000-0001-9752-2531)
Source: Applied Cognitive Psychology. 2025 39(3).
Availability: Wiley. Available from: John Wiley & Sons, Inc. 111 River Street, Hoboken, NJ 07030. Tel: 800-835-6770; e-mail: cs-journals@wiley.com; Web site: https://www.wiley.com/en-us
Peer Reviewed: Y
Page Count: 16
Publication Date: 2025
Document Type: Journal Articles
Reports - Research
Descriptors: Concept Teaching, Concept Formation, Learning Processes, Schemata (Cognition), Imagery, Sentences, Semantics, Priming, Recall (Psychology), Familiarity, Difficulty Level, Instructional Design, Instructional Materials
DOI: 10.1002/acp.70054
ISSN: 0888-4080
1099-0720
Abstract: The coherence principle suggests removing unnecessary--or seductive--content from educational texts to reduce cognitive load. However, the binary proposition that all seductive details should be excluded neglects images' potential to prime semantically related concepts, which makes texts easier to process. It was hypothesized that this priming would cause at least tangentially related images to enhance processing and recall of concepts. Participants learned 24 concepts under four conditions: direct depictions, tangentially related and unrelated images, and no image. Participants' fixation durations on concepts, their complementing sentences and images, and recall performance were measured. Multilevel models revealed that coherence effects were only present for unrelated images and that images that are at least tangentially related facilitated learning. These effects were unaffected by participants' familiarity with concepts. The study concludes that semantically related images may outweigh their cognitive load, suggesting that educators should consider their priming potential when designing instructional materials.
Abstractor: As Provided
Notes: https://osf.io/evzsw/?view_only=527ead2d91c0414192f25131199fa519
Entry Date: 2025
Accession Number: EJ1474588
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGf1Pb7urGJwklGuLJUzretAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDDbEyrHjgxiQmyf9XQIBEICBmyxpiE_hfvO_lLn4dh5_2LgBHylWx-twcpDQP2g8gNGhw-UHbcNMqX5AgkAcjlr7OThD4cVjDTzRV1uEWQ84OcRU-BYV_nGvIEW4ISeTwjGQJupjMkwSPhNWQRf9Elm8ud9mW015rRhyR55djIGYb7R-TX49SqV4jBORkHuX2BQAy8_QpT_vRZv6ws2xXhK9Ssi9zGnwHUzpqHsC
Text:
  Availability: 1
  Value: <anid>AN0186111736;bu801may.25;2025Jun25.05:59;v2.2.500</anid> <title id="AN0186111736-1">Unraveling Coherence: The Impact of Image‐Relatedness in Concept Learning </title> <p>The coherence principle suggests removing unnecessary—or seductive—content from educational texts to reduce cognitive load. However, the binary proposition that all seductive details should be excluded neglects images' potential to prime semantically related concepts, which makes texts easier to process. It was hypothesized that this priming would cause at least tangentially related images to enhance processing and recall of concepts. Participants learned 24 concepts under four conditions: direct depictions, tangentially related and unrelated images, and no image. Participants' fixation durations on concepts, their complementing sentences and images, and recall performance were measured. Multilevel models revealed that coherence effects were only present for unrelated images and that images that are at least tangentially related facilitated learning. These effects were unaffected by participants' familiarity with concepts. The study concludes that semantically related images may outweigh their cognitive load, suggesting that educators should consider their priming potential when designing instructional materials.</p> <p>Keywords: activation theory; coherence principle; decorative pictures; seductive details; semantic priming; tangentially related images</p> <hd id="AN0186111736-2">Introduction</hd> <p>Imagine an instructional text intended to inform learners about how sound waves work, accompanied by an image of a concert. Predominant research suggests that it would be better to exclude such an image (e.g., Rey [<reflink idref="bib42" id="ref1">42</reflink>]). This is because, despite its aesthetic appeal, the image does little to clarify the properties of sound waves and requires unnecessary effort to process.</p> <p>Most literature on this topic tends to compare the inclusion of these irrelevant images to their exclusion or contrasts them with relevant images (Korbach et al. [<reflink idref="bib21" id="ref2">21</reflink>]; Sundararajan and Adesope [<reflink idref="bib53" id="ref3">53</reflink>]). In reality, visual images may vary across a range of dimensions, making the binary classification of relevant versus irrelevant overly simplistic. In particular, images can be categorized based on the degree to which they contribute to an instructional objective (relevance), their semantic connection to the instructional content (relatedness), and their educational purpose (instructional or decorative).</p> <p>In this light, the image of a concert in an explanatory text about sound waves may serve merely a decorative purpose and may be considered irrelevant. However, it does have a semantic relation to sound waves. Research suggests that this semantic relatedness can facilitate cognitive processing, raising intriguing questions about the potential impact of such images, even when they do not directly contribute to the instructional objective (Angwin et al. [<reflink idref="bib4" id="ref4">4</reflink>]).</p> <p>By acknowledging both the mental effort associated with processing visual images and the potential of semantically related images to facilitate the learning of concepts, the present study offers a nuanced approach to multimedia learning. It proposes that visual images, when at least remotely related to the instructional content, could potentially reduce the mental effort needed for processing, possibly outweighing the additional effort associated with processing the image. The present study categorizes the relatedness of visual images to presented concepts and assesses its impact on learning. In doing so, the study aims to refine instructional design guidelines that suggest uniformly excluding images that are not completely related (Rey [<reflink idref="bib42" id="ref5">42</reflink>]).</p> <hd id="AN0186111736-3">Images in Instructional Texts</hd> <p>Research on the use of visual images in instructional texts has predominantly focused on instructional pictures, which serve primarily an educational purpose (Lenzner et al. [<reflink idref="bib25" id="ref6">25</reflink>]). These images are often included to aid learners in grasping the physical resemblance, logical structures, or functional analogies of concepts. Educational research has consistently shown that learners achieve superior learning outcomes when textual information is complemented by instructional pictures (e.g., Hu et al. [<reflink idref="bib18" id="ref7">18</reflink>]; Levie and Lentz [<reflink idref="bib26" id="ref8">26</reflink>]; Levin et al. [<reflink idref="bib27" id="ref9">27</reflink>]; Lindner [<reflink idref="bib28" id="ref10">28</reflink>]; Lindner et al. [<reflink idref="bib29" id="ref11">29</reflink>]; Mayer [<reflink idref="bib32" id="ref12">32</reflink>]), demonstrating that such visuals can improve retention, deepen comprehension of complex subjects, and enhance problem‐solving skills (Mayer [<reflink idref="bib30" id="ref13">30</reflink>], [<reflink idref="bib32" id="ref14">32</reflink>]; Moreno and Mayer [<reflink idref="bib36" id="ref15">36</reflink>]). Mayer introduced the term "multimedia effect" to describe this phenomenon, where learners benefit from the combined use of text and images.</p> <p>In addition to instructional pictures, educational texts often include decorative pictures, intended to make the material aesthetically pleasing and engaging, which typically have less semantic relation to the text (Carney and Levin [<reflink idref="bib9" id="ref16">9</reflink>]). The key distinction between instructional and decorative pictures lies in their primary function: instructional pictures are informative, whereas decorative pictures aim to enhance the aesthetic experience. Although images are usually included with one of these two purposes, earlier multimedia research already provided a more nuanced perspective. Takahashi ([<reflink idref="bib55" id="ref17">55</reflink>]) described these functions as existing along a two‐dimensional spectrum with a decorative and instructional dimension. This suggests that images can be more decoratively or more instructionally oriented. However, it also suggests that decoratively oriented images can hold instructional relevance, depending on factors such as their coherence with the instructional content.</p> <hd id="AN0186111736-4">Coherence Principle</hd> <p>An important condition for images to enhance learning (i.e., for the multimedia effect to occur) is coherence between the image and the text (Carney and Levin [<reflink idref="bib9" id="ref18">9</reflink>]). Mayer's coherence principle (2005) emphasizes that individuals learn better when extraneous content, such as decorative pictures, is excluded from instructional materials. It serves as a broader framework that explains the impact of seductive details—a term used to describe irrelevant information added to learning materials, usually with the intent of making them more interesting (Garner et al. [<reflink idref="bib14" id="ref19">14</reflink>]; Mayer [<reflink idref="bib31" id="ref20">31</reflink>]). Where the coherence principle suggests excluding <emph>any</emph> irrelevant information, research on seductive details usually focuses on irrelevant information that is at least tangentially related to the material (e.g., Bender et al. [<reflink idref="bib5" id="ref21">5</reflink>]). Being frequently included in instructional materials for aesthetic purposes, decorative images serve as a prominent example of seductive details (Sung and Mayer [<reflink idref="bib54" id="ref22">54</reflink>]).</p> <p>According to the coherence principle, seductive details occupy a portion of the working memory, which results in additional cognitive load (Mayer [<reflink idref="bib31" id="ref23">31</reflink>]), and divert learners' attention away from processing relevant information (Lehman et al. [<reflink idref="bib24" id="ref24">24</reflink>]). An alternative explanation for the negative impact of seductive details on learning is the schema interference hypothesis. This hypothesis posits that seductive details interfere with the learning processes by priming inappropriate schemas, hence wasting cognitive capacity and cluttering working memory with unrelated retrieved knowledge (Rey [<reflink idref="bib42" id="ref25">42</reflink>]).</p> <p>Echoing the diversity of explanations for the seductive details effect, research regarding the effect of the coherence principle remains inconclusive as well. Various findings indicate both negative, positive, and neutral effects of irrelevant information in learning materials (Korbach et al. [<reflink idref="bib21" id="ref26">21</reflink>], [<reflink idref="bib22" id="ref27">22</reflink>]; Mayer [<reflink idref="bib31" id="ref28">31</reflink>]; Rey [<reflink idref="bib42" id="ref29">42</reflink>]; Scheiter et al. [<reflink idref="bib50" id="ref30">50</reflink>]; Wang and Adesope [<reflink idref="bib61" id="ref31">61</reflink>]). The differences in these findings may result from the variety of aspects believed to underpin the negative impact of irrelevant information. For instance, from a cognitive load perspective, these coherence effects might vary due to differences in format, saliency, or unit density (Wang et al. [<reflink idref="bib62" id="ref32">62</reflink>]). Alternatively, the varying impact of seductive details and other irrelevant material may be explained by the dissimilarities in the relatedness of the material being learned, which leads to diverse activation of (interfering) information in long‐term memory (Anderson [<reflink idref="bib3" id="ref33">3</reflink>]).</p> <hd id="AN0186111736-5">Activation Theory</hd> <p>Activation theory suggests that when a node of a concept is activated in the brain, the activation spreads to other related nodes in the brain, which can lead to easier access to information (Anderson [<reflink idref="bib3" id="ref34">3</reflink>]). Neely ([<reflink idref="bib37" id="ref35">37</reflink>]) related the activation process to semantic priming, which suggests that the processing of a word is easier when related concepts are shown in addition to that word (Meyer and Schvaneveldt [<reflink idref="bib35" id="ref36">35</reflink>]; Tulving and Schacter [<reflink idref="bib59" id="ref37">59</reflink>]). Following this reasoning, visual images may activate a diverse array of semantically related concepts, consequently reducing the mental effort necessary to process a key concept. Neuroimaging studies using functional Magnetic Resonance Imaging (fMRI) have supported this by showing that, when processing concepts, exposure to semantically related concepts reduces neural activity in regions associated with semantic processing, such as the left inferior prefrontal cortex (Heath et al. [<reflink idref="bib15" id="ref38">15</reflink>]; Tivarus et al. [<reflink idref="bib56" id="ref39">56</reflink>]). This suggests a reduction in mental effort necessary for processing concepts that have been activated recently, either through text or images (Kircher et al. [<reflink idref="bib20" id="ref40">20</reflink>]; Wagner et al. [<reflink idref="bib60" id="ref41">60</reflink>]). In theory, presenting a tangentially related image may already yield such cognitive processing benefits compared to presenting an unrelated image or no image at all. However, this reduction in mental effort may be compensated by the extraneous cognitive load introduced by the visual elements.</p> <hd id="AN0186111736-6">Familiarity</hd> <p>Furthermore, recent findings indicate that semantic priming effects are influenced by repetition suppression; a reduction in neural activity after repeated exposure (Kim [<reflink idref="bib19" id="ref42">19</reflink>]). This suggests that individuals' mental effort is reduced when processing concepts that they have been repeatedly exposed to, or in other words, familiar concepts. Highly familiar concepts are easier to access and retrieve from long‐term memory (Ratcliff and McKoon [<reflink idref="bib39" id="ref43">39</reflink>]), and their associations with other concepts tend to be stronger as they are better integrated in memory (Anaki and Henik [<reflink idref="bib2" id="ref44">2</reflink>]). This explains the reduction in neural activity when processing concepts that are more familiar (Kim [<reflink idref="bib19" id="ref45">19</reflink>]).</p> <p>While stronger connections to other concepts make the priming effects on familiar concepts stronger, familiar concepts are also easier to retrieve by themselves, and thus benefit less from the priming effect. Conversely, completely unfamiliar images are not yet stored in memory and can thus not be primed. Therefore, the semantic priming benefits might be strongest if concepts are familiar but not too well‐known. Varying degrees of familiarity have also been proposed as a possible explanation for the contrasting results in seductive details research (Rey [<reflink idref="bib42" id="ref46">42</reflink>]).</p> <hd id="AN0186111736-7">Coherence Versus Semantic Priming Effects</hd> <p>Building on these frameworks, two opposing effects come into play with respect to irrelevant images. On the one hand, such images increase cognitive load, as additional effort is required to process material that lacks coherence with the instructional content (i.e., the coherence effect). On the other hand, irrelevant images may also reduce the effort needed to process learning material if they possess some degree of relatedness to the content (i.e., the semantic priming effect). Since the degree of relatedness of images to the text varies, the reduction in cognitive load from semantic priming can either outweigh or be outweighed by the additional cognitive load imposed by the image. Theoretically, there would exist a breakeven point where these semantic priming and coherence effects are equal. Beyond this point, images with stronger semantic relations would actively aid in processing and enhance learning.</p> <p>While the semantic priming effects vary alongside images' relatedness (Neely [<reflink idref="bib37" id="ref47">37</reflink>]), coherence effects are dependent on factors like the images' size, unit density (e.g., three cats instead of one) and their saliency (Wang et al. [<reflink idref="bib62" id="ref48">62</reflink>]). Together, all these properties determine the balance between the semantic priming and coherence effects that are associated with the image. Variance in these properties would possibly explain the inconclusiveness of research on irrelevant visual images (Rey [<reflink idref="bib42" id="ref49">42</reflink>]; Sundararajan and Adesope [<reflink idref="bib53" id="ref50">53</reflink>]; Towler et al. [<reflink idref="bib58" id="ref51">58</reflink>]) as they determine the interplay between the coherence and semantic priming effects.</p> <hd id="AN0186111736-8">Non‐Instructional Potential of Images</hd> <p>Images' potential to activate semantically related material has recently been studied by Scherer et al. ([<reflink idref="bib51" id="ref52">51</reflink>]), who found learning‐enhancing potential for images without an instructional purpose. They examined decorative images that were closely related to the to‐be‐learned material and found that their inclusion enhanced recall performance. However, the degree of relatedness required for such benefits remains unclear. In particular, it is still uncertain whether images with weaker semantic connections to the instructional content (i.e., tangentially related images) can provide similar advantages. This calls for further research to quantify the effect and determine the threshold of relatedness necessary to yield cognitive benefits.</p> <p>Addressing this gap could reveal instances where including images deemed irrelevant may still be worthwhile. On a practical note, a nuanced perspective on images that do not directly contribute to the instructional objective is also valuable because of their potential to enhance learning in an indirect way. For instance, decoratively oriented images can have a mood‐enhancing effect, promoting calmness or increased alertness, which can positively impact learning (Lenzner et al. [<reflink idref="bib25" id="ref53">25</reflink>]; Schneider et al. [<reflink idref="bib52" id="ref54">52</reflink>]). Furthermore, decorative images can reduce the perceived difficulty of the learning material, thereby lowering anxiety and helping learners reach optimal arousal levels (Alarcão and Fonseca [<reflink idref="bib1" id="ref55">1</reflink>]; Lenzner et al. [<reflink idref="bib25" id="ref56">25</reflink>]; Schneider et al. [<reflink idref="bib52" id="ref57">52</reflink>]).</p> <hd id="AN0186111736-9">Eye‐Tracking</hd> <p>Eye‐tracking technology is commonly used to study the processing of pictures (e.g., Scharinger et al. [<reflink idref="bib49" id="ref58">49</reflink>]). Eye‐tracking technology is optimal for measuring text processing speed due to its ability to capture detailed and precise data on eye movements, revealing real‐time processing and comprehension efforts (Rayner [<reflink idref="bib40" id="ref59">40</reflink>]). Measuring fixation durations on specific words and images provides insights into how different types of images impact the processing of texts. It also reveals how much individuals look at images and how this relates to learning outcomes.</p> <hd id="AN0186111736-10">The Present Study</hd> <p>The present study aims to provide a more nuanced perspective on the incorporation of visual images within instructional texts. To investigate the interplay between semantic priming and coherence effects, it categorizes visual images into three distinct conditions: direct depictions of a concept, tangentially related images, and unrelated images (for instance, for the word "eclipse", image conditions were the moon overshadowing the sun, a space shuttle, or a carnivorous plant, respectively). In addition, a fourth condition is included that does not contain images.</p> <p>Including visual images introduces extraneous cognitive load in the learning task. For (at least tangentially) related images, this cognitive load may be compensated by the process of semantic priming, which reduces the mental effort necessary for processing. Unrelated images, on the other hand, may not yield such semantic priming potential, and none of their cognitive load can therefore be compensated, leading to impaired learning. In the absence of images, no semantic priming effects and no additional cognitive load are imposed, making this condition serve as a concrete baseline.</p> <p>Therefore, the present study posits that the direct depiction and tangentially related image conditions will yield improved recall performance and faster processing of concepts compared to the no‐image condition. Moreover, it anticipates that all these conditions will result in better recall performance and faster processing than the condition with completely unrelated images. Finally, it is anticipated that the direct depiction condition will outperform the tangentially related condition, as the semantic priming effect is expected to be stronger when the image closely aligns with the core concept. This leads to the hypothesized ranking of conditions as follows: (Direct depiction>Tangentially related>No image>Unrelated image).</p> <p>This hypothesized structure was measured across four main outcome variables: the total duration a participant fixated on (<reflink idref="bib1" id="ref60">1</reflink>) the concept and (<reflink idref="bib2" id="ref61">2</reflink>) its surrounding sentence, (<reflink idref="bib3" id="ref62">3</reflink>) the total duration participants spent on the specific trial, and (<reflink idref="bib4" id="ref63">4</reflink>) recall performance (i.e., participants' scores on a test of the concepts' definitions).</p> <p>The total fixation duration on concepts and surrounding text provides an estimate of how images impact the processing of textual information, making them concrete measures of processing efficiency (Rayner [<reflink idref="bib40" id="ref64">40</reflink>]). The priming effects were hypothesized to be most pronounced for fixations on the concepts, as this is the part of the text that is targeted to be primed with the image. This would result in a shorter total fixation duration on concepts for the tangentially related and direct depiction conditions compared to the no‐image and unrelated image conditions. For total fixation durations on the surrounding sentences, this applies as well, albeit to a lesser degree.</p> <p>The trial duration measure also reflects the time participants were distracted, providing a relative measure of processing efficiency. Since viewing images extends the total trial duration, this measure inherently penalizes the act of looking at an image, unlike the fixation measures. Consequently, this metric may reveal smaller or no benefits of tangentially related images and direct depictions compared to the no‐image condition. Lastly, effects on recall performance are also expected, as the related images may enhance the integration of concepts within established mental representations (Anderson [<reflink idref="bib3" id="ref65">3</reflink>]), thereby facilitating easier recall of their definitions.</p> <p>On the one hand, for semantic priming effects to occur and aid learning, words must be available in the individual's long‐term memory, as unknown concepts would not be represented in memory. On the other hand, high familiarity would allow rapid access to the concepts in memory anyway, making priming unnecessary. Semantic priming effects are likely to be strongest for concepts that are moderately familiar to an individual (i.e., neither very well‐known nor unknown). Therefore, familiarity measures and nonlinear interaction effects were included as covariates to examine and control for potential nonlinear interactions.</p> <hd id="AN0186111736-11">Methods</hd> <p></p> <hd id="AN0186111736-12">Participants and Design</hd> <p>The sample for the study comprised 31 Dutch university students, with 22 students majoring in social sciences, four in natural sciences, two in environmental sciences, and one each in history, finance, engineering, and IT. These students (11 males and 20 females) were aged between 18 and 26 years (<emph>M</emph> = 21.77, SD = 1.78) and all had normal or corrected‐to‐normal vision.</p> <p>The study employed a within‐subjects design. Each participant was exposed to all four image conditions and all 24 concepts. Participants were tested six times in each image condition (i.e., no images, unrelated images, tangentially related images, and direct depictions).</p> <p>The data follows a cross‐classified, multilevel structure. McNeish and Stapleton ([<reflink idref="bib34" id="ref66">34</reflink>]) recommend at least 30 classes (individuals, in the present study) to minimize bias in analyses, particularly when focusing on level 1 predictors. With 24 concepts tested, the final sample of 31 participants provided 744 cases, meeting power criteria for predicting level 1 effects and variances (Hox et al. [<reflink idref="bib17" id="ref67">17</reflink>]).</p> <hd id="AN0186111736-13">Materials</hd> <p></p> <hd id="AN0186111736-14">Concept Learning Task</hd> <p>The concept learning task was divided into three blocks. In each block, participants were presented with eight trials, for a total of 24 trials. Each trial included a concept and its definition, intended to be learned by participants. The concepts and definitions were (translated) rephrases of (digital) glossaries in Dutch.</p> <p>Most trials paired the presented concept with an image: a direct depiction of the concept word, a tangentially related image, or an unrelated image, while control trials did not contain an image. Each trial could belong to one of these four conditions (see Appendix Table A1), and each block contained two trials of each condition. A trial example is shown in Figure 1.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01may25/acp70054-fig-0001.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70054-fig-0001.jpg" title="1 Four conditions of trial "eclipse". The presented trial is an English translation of the originally Dutch trial. Direct depiction on the upper left, tangentially related image on the upper right, non‐related image on the lower left, and control trial on the lower right. AOIs in the trials were positioned around the image, sentence, and the concept (here the word "eclipse"). The concept and sentence AOIs included an additional 11.2 mm margin for fixation (1.1° horizontal, 1.1° vertical). Images referenced as follows: The eclipse by Edwin Remsberg (from Getty Images), the space shuttle by Dima Zel (from Getty Images), and the Venus fly trap from Encyclopaedia Britannica." /> </p> <p></p> <p>The degree of relatedness was operationally defined in terms of semantic distance (e.g., Rips et al. [<reflink idref="bib43" id="ref68">43</reflink>]), using intermediate concepts to indicate varying relatedness. For direct depictions, the images represented the concept itself. Tangentially related images were mediated by one or two intermediate concepts (e.g., <emph>shuttle → space → celestial bodies→ eclipse</emph>). Unrelated images were either connected through more than three intermediate concepts or lacked any meaningful connection (e.g., <emph>Venus flytrap ⊗ eclipse</emph>).</p> <p>All three possible images attached to a respective concept were selected to be similar in color (distribution) and the presence of faces (Beymer et al. [<reflink idref="bib6" id="ref69">6</reflink>]). All images had a resolution of 612 × 468 pixels and covered a visual angle of 14.6° horizontally and 11.8° vertically. Images were selected to not serve a specific instructional purpose (e.g., explaining the concept).</p> <p>There were four versions of the concept learning task. Each version of the experiment ordered trials in each block differently and paired the trials with different conditions. This was done in a mutually exclusive manner, meaning that each condition was paired with each concept in one of the versions. To ensure an equal number of participants per version, participants were distributed among the versions through a counterbalanced approach.</p> <p>Counterbalancing each concept across the conditions ensured that differences in conditions were not confounded by potential differences in concept difficulty. For instance, if all "eclipse" trials were in the direct depiction condition, an easier "eclipse" trial could create the false impression that this condition performed better.</p> <hd id="AN0186111736-16">Familiarity Assessment</hd> <p>Familiarity with each presented concept was assessed at the end of the experiment using a 10‐point scale, ranging from completely unknown (<reflink idref="bib1" id="ref70">1</reflink>) to well‐known (<reflink idref="bib10" id="ref71">10</reflink>). To ensure that the experimental manipulations did not influence the familiarity ratings, participants were specifically asked to recall how familiar the concepts were to them prior to the start of the experiment.</p> <p>The familiarity variable was also used to assess the criterion validity of the concept learning task. As previous research has consistently found moderate to small relationships between word familiarity and fixation durations, significant main effects of the familiarity variable in the models were considered indicative of criterion validity (Lahoud et al. [<reflink idref="bib23" id="ref72">23</reflink>]; Reichle et al. [<reflink idref="bib41" id="ref73">41</reflink>]; Williams and Morris [<reflink idref="bib63" id="ref74">63</reflink>]).</p> <hd id="AN0186111736-17">Recall Test</hd> <p>At the end of each block, participants completed a recall test on the eight concepts they learned in that block. In this test, respondents were tasked with providing the definitions of the concepts. A 6‐min countdown allowed respondents to estimate the amount of time they should allocate to the test, though they were not prohibited from exceeding the time limit. The scoring system for this test allowed for either zero, one, or two points per concept, with a maximum of two points obtainable if both predefined criteria for a definition were met and one point if just one of the two criteria for a definition was met.</p> <p>Inter‐rater reliability was considered through detailed criteria for the number of points to be obtained in each question. To ensure consistency, a pilot respondent's test was graded jointly by two researchers. Then, both researchers individually graded all tests, achieving a quadratic weighted <emph>κ</emph> = 0.57, indicating moderate agreement (McHugh [<reflink idref="bib33" id="ref75">33</reflink>]). Instances where the researchers' grades differed were discussed to reach a consensus. These agreed‐upon values were used as the final test scores.</p> <hd id="AN0186111736-18">Procedure</hd> <p>Prior to the experiment, participants were instructed to read the digital information letter, which was sent in advance and presented during the session. Additionally, participants were asked to complete the digital informed consent form during the experiment. By adhering to ethical guidelines, the study obtained approval from Utrecht University's Faculty Ethical Review Board (FERB) as part of an existing research line.</p> <p>The experiment began with the participant entering the soundproof room where the eye‐tracker was set up. After the setup, the participant was left alone, and instruction slides, along with one example trial, were presented. Each block started with a nine‐point calibration process, including a four‐point validation procedure. In each block, participants were instructed to study eight concepts and complete a recall test afterward. The concept learning task was self‐paced, with participants utilizing the space bar to progress. To standardize the initiation point for each trial and mitigate biases, a fixation cross was displayed in the upper center of the screen for one second before the presentation of each trial. After completing a block of the concept learning task, participants were given an optional one‐minute break before proceeding to the recall task, which was followed by a new block of the concept learning task. After finishing the experiment, all participants completed a brief demographic survey (including gender, age, and field of study). The entire experiment lasted approximately 30–45 min.</p> <hd id="AN0186111736-19">Eye Tracking Setup and Data Collection</hd> <p>Eye‐tracking data were collected using a Tobii Pro Spectrum (TPSP1‐010200527044) (Tobii [<reflink idref="bib57" id="ref76">57</reflink>]), with the standard Tobii I‐VT filter (with fixations being defined as having a minimal duration of 60 ms and a maximum speed of 30 deg/s). Stimuli were presented on an HP E233 monitor with a presentation resolution of 1920 × 1080 pixels (23.3 in.). Eye movements were recorded with a sampling rate of 300 Hz and a viewing distance of approximately 59 cm. A chinrest was employed to stabilize participants' head position, for which Tobii Lab Pro software was used to determine the optimal position.</p> <p>In each trial, three Areas of Interest (AOI) were incorporated around the concept itself (i.e., keyword), the entire sentence, and the presented image with a 1.1° margin around the text and image. For trials in the no‐image condition, (empty) AOIs were incorporated in the exact same locations as in the image‐containing conditions. For each AOI, the total fixation duration (i.e., the sum of all fixation durations inside the AOI) was calculated. Trial duration, defined as the time between trial onset and the participant's initiation to proceed, was also measured.</p> <p>For eye‐tracking data quality, a threshold for accuracy and precision of less than 1.1° was set as a quality criterion, based on additional AOI space around the keywords and sentences. All observations within a block were removed if the accuracy or precision measures exceeded this threshold.</p> <hd id="AN0186111736-20">Data Analysis</hd> <p>The variables in the main analyses comprised total keyword and sentence fixation durations, trial duration, and recall score. To gain further insight into the main analyses' findings, an additional outcome variable, image fixation duration, was also investigated.</p> <p>The main analyses comprised cross‐classified multilevel analyses, with the first level encompassing individual trials while the second level included concept words and respondents. As recall scores were ordinal, a multilevel cumulative link model (Christensen [<reflink idref="bib10" id="ref77">10</reflink>]) with Laplace approximation (Hedeker [<reflink idref="bib16" id="ref78">16</reflink>]) was instead conducted for this variable. For all outcome variables, a stepwise approach was adopted for model building. First, a baseline model that included only the condition variables was fitted. Second, the familiarity variable was added to the model. Third, a quadratic term for familiarity and interactions with the conditions were added. If the addition of the familiarity variable or its quadratic interactions significantly improved the model fit (<emph>α</emph> < 0.05), they were included as covariates in subsequent analyses.</p> <p>The metrics were designed to accurately reflect meaningful engagement with the text and concepts. To ensure that the data were not influenced by trials in which participants accidentally skipped the text, fixation durations that were too short to indicate true cognitive processing were excluded. This exclusion was based on reported reading speeds of proficient readers (Brysbaert [<reflink idref="bib8" id="ref79">8</reflink>]), ensuring that no genuine values would be lost. Consequently, any trial with fixation durations below a minimum threshold was excluded from the analysis (i.e., ≥ 150 ms for keyword fixation durations; ≥ 1500 ms for sentence fixation and trial durations). Such values were encountered in 0.4% of the keyword fixation durations, 0.3% of sentence fixation durations, and none of the trial durations.</p> <p>Inspection of histograms and Q‐Q plots (observed residuals against expected normal values) revealed no severe violations of (multivariate) normality. Outlier detection was completed using trimmed means and trimmed standard deviations (Elamir and Seheult [<reflink idref="bib13" id="ref80">13</reflink>]). The lower 2.5th percentile and the upper 97.5th percentile were trimmed from these calculations, and values deviating over three trimmed standard deviations from the trimmed mean were excluded from any further analyses. Such outliers were observed in 4.5% of the keyword fixations, 7.5% of the sentence fixations, and 9.7% of the trial durations. After removing outliers, 652 out of the initial 736 observations remained in the sample. The results of the analyses were directionally similar when conducted with and without outlier removal. However, the effects were reduced in magnitude compared to the analyses in which outliers were deleted. The results reported are for the analyses with outliers removed.</p> <p>Standardized regression coefficients were used as effect sizes. Following the guidelines proposed by Cohen ([<reflink idref="bib11" id="ref81">11</reflink>]), values of 0.10 were considered small, 0.30 were considered medium, and 0.50 were considered large.</p> <hd id="AN0186111736-21">Results</hd> <p>The dataset is made available on OSF[<reflink idref="bib1" id="ref82">1</reflink>]. Descriptive statistics on accuracy, precision, and data loss scores are presented in Table 1. Almost no participant displayed accuracy or precision values exceeding 1.1° or a proportion of data losses exceeding 0.30, except for a single block from one participant with a precision standard deviation of 1.34, which was excluded from further analysis.</p> <p>1 TABLE Eye tracking data quality.</p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left">Measurement</th><th align="center">Mean</th><th align="center">SD</th><th align="center">Min</th><th align="center">Max</th></tr></thead><tbody valign="top"><tr><td align="left">Accuracy</td><td align="center">0.39°</td><td align="center">0.16°</td><td align="center">0.13°</td><td align="center">0.90°</td></tr><tr><td align="left">Precision (SD)</td><td align="center">0.26°</td><td align="center">0.23°</td><td align="center">0.05°</td><td align="center">0.95°</td></tr><tr><td align="left">Precision (RMS)</td><td align="center">0.14°</td><td align="center">0.13°</td><td align="center">0.03°</td><td align="center">0.83°</td></tr><tr><td align="left">Proportion data loss</td><td align="center">0.04</td><td align="center">0.05</td><td align="center">0.00</td><td align="center">0.19</td></tr></tbody></table> </ephtml> </p> <p>1 <emph>Note:n</emph> = 92 (3 calibration steps, 31 participants). SD refers to the precision's standard deviation. RMS refers to the precision's root mean square of the precision.</p> <p>Descriptive statistics are presented in Table 2. The results of the analyses, including estimates of familiarity and condition effects for each outcome variable, are shown in Table 3, where image conditions are compared to the no‐image condition. Table 4 displays pairwise comparisons between the image conditions in matrix form.</p> <p>2 TABLE Descriptive statistics.</p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left" /><th align="center">Direct depiction</th><th align="center">Tangentially related image</th><th align="center">Unrelated image</th><th align="center">No image</th><th align="center">Overall</th></tr><tr><th align="center"><italic>M</italic></th><th align="center">SD</th><th align="center"><italic>M</italic></th><th align="center">SD</th><th align="center"><italic>M</italic></th><th align="center">SD</th><th align="center"><italic>M</italic></th><th align="center">SD</th><th align="center"><italic>M</italic></th><th align="center">SD</th></tr></thead><tbody valign="top"><tr><td align="left">Keyword fixation duration</td><td align="center">1.79</td><td align="center">1.13</td><td align="center">2.07</td><td align="center">1.34</td><td align="center">2.33</td><td align="center">1.28</td><td align="center">2.43</td><td align="center">1.53</td><td align="center">2.16</td><td align="center">1.35</td></tr><tr><td align="left">Sentence fixation duration</td><td align="center">8.10</td><td align="center">4.36</td><td align="center">8.73</td><td align="center">4.40</td><td align="center">9.65</td><td align="center">4.63</td><td align="center">9.64</td><td align="center">4.64</td><td align="center">9.03</td><td align="center">4.55</td></tr><tr><td align="left">Trial duration</td><td align="center">12.09</td><td align="center">6.20</td><td align="center">12.32</td><td align="center">5.79</td><td align="center">13.46</td><td align="center">6.10</td><td align="center">12.16</td><td align="center">5.99</td><td align="center">12.50</td><td align="center">6.04</td></tr><tr><td align="left">Recall score (0–2)</td><td align="center">1.71</td><td align="center">0.56</td><td align="center">1.63</td><td align="center">0.69</td><td align="center">1.35</td><td align="center">0.81</td><td align="center">1.49</td><td align="center">0.76</td><td align="center">1.55</td><td align="center">0.72</td></tr><tr><td align="left">Image fixation duration</td><td align="center">1.46</td><td align="center">1.71</td><td align="center">1.02</td><td align="center">1.03</td><td align="center">0.98</td><td align="center">1.15</td><td align="center">0.03</td><td align="center">0.32</td><td align="center">0.87</td><td align="center">1.27</td></tr><tr><td align="left">Familiarity (1–10)</td><td align="center">5.48</td><td align="center">3.37</td><td align="center">5.26</td><td align="center">3.41</td><td align="center">5.45</td><td align="center">3.41</td><td align="center">4.89</td><td align="center">3.27</td><td align="center">5.27</td><td align="center">3.36</td></tr></tbody></table> </ephtml> </p> <ulist> <item>2 <emph>Note:n =</emph> 652, 31 individuals and 24 keywords. Fixation and trial durations are presented in seconds.</item> <item>3 TABLE Multilevel results per outcome variable.</item> </ulist> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left" /><th align="center">Keyword fixation</th><th align="center">Sentence fixation</th><th align="center">Trial duration</th><th align="center">Recall score</th><th align="center">Image fixation</th></tr></thead><tbody valign="top"><tr><td align="left">Fixed part</td></tr><tr><td align="left">Intercept</td><td align="center">3108.94<xref ref-type="fn" rid="tfn6" />(185.69)0.05</td><td align="center">11865.17<xref ref-type="fn" rid="tfn6" />(802.46)0.14</td><td align="center">15532.60<xref ref-type="fn" rid="tfn6" />(1067.93)0.16</td><td align="center" /><td align="center">427.86<xref ref-type="fn" rid="tfn5" />(146.31)0.05</td></tr><tr><td align="left">Direct depiction</td><td align="center">−580.22<xref ref-type="fn" rid="tfn6" />(117.16)−0.19</td><td align="center">−1302.57<xref ref-type="fn" rid="tfn6" />(332.10)−0.12</td><td align="center">174.63(432.85)0.01</td><td align="center">0.70<xref ref-type="fn" rid="tfn5" />(0.26)0.30</td><td align="center">1443.05<xref ref-type="fn" rid="tfn6" />(113.50)0.49</td></tr><tr><td align="left">Tangentially related image</td><td align="center">−301.19<xref ref-type="fn" rid="tfn4" />(117.34)−0.10</td><td align="center">−564.35 (332.33)−0.05</td><td align="center">577.21(433.16)0.04</td><td align="center">0.54<xref ref-type="fn" rid="tfn4" />(0.26)0.23</td><td align="center">988.22<xref ref-type="fn" rid="tfn6" />(113.87)0.33</td></tr><tr><td align="left">Unrelated image</td><td align="center">−56.34(117.32)−0.02</td><td align="center">236.86(332.30)0.02</td><td align="center">1600.06<xref ref-type="fn" rid="tfn6" />(433.12)0.11</td><td align="center">−0.51<xref ref-type="fn" rid="tfn4" />(0.24)−0.22</td><td align="center">985.64<xref ref-type="fn" rid="tfn6" />(113.84)0.33</td></tr><tr><td align="left">Familiarity</td><td align="center">−122.49<xref ref-type="fn" rid="tfn6" />(16.19)−0.31</td><td align="center">−340.64<xref ref-type="fn" rid="tfn6" />(48.41)−0.25</td><td align="center">−507.94<xref ref-type="fn" rid="tfn6" />(62.76)−0.28</td><td align="center">0.11<xref ref-type="fn" rid="tfn6" />(0.03)0.41</td><td align="center">−64.71<xref ref-type="fn" rid="tfn6" />(12.76)−0.17</td></tr><tr><td align="left">Random part</td></tr><tr><td align="left">σ<sup>2</sup><sub>e</sub></td><td align="center">1,114,344</td><td align="center">8,926,124</td><td align="center">15,164,356</td><td align="center" /><td align="center">1,052,548</td></tr><tr><td align="left">σ<sup>2</sup><sub>u (Individual)</sub></td><td align="center">458,125</td><td align="center">11,237,586</td><td align="center">22,200,761</td><td align="center">0.78</td><td align="center">338,696</td></tr><tr><td align="left">σ<sup>2</sup><sub>u (Concept)</sub></td><td align="center">158,732</td><td align="center">4,097,615</td><td align="center">5,692,319</td><td align="center">0.27</td><td align="center">2172</td></tr><tr><td align="left">Deviance</td><td align="center">11034.4</td><td align="center">12446.7</td><td align="center">12792.4</td><td align="center" /><td align="center">10955.2</td></tr><tr><td align="left">AIC</td><td align="center">11050.4</td><td align="center">12462.7</td><td align="center">12808.4</td><td align="center">1016.42</td><td align="center">10971.2</td></tr><tr><td align="left">BIC</td><td align="center">11086.3</td><td align="center">12498.5</td><td align="center">12844.2</td><td align="center">989.453</td><td align="center">11007.1</td></tr></tbody></table> </ephtml> </p> <ulist> <item>3 <emph>Note:n</emph> = 652 observations, 31 individuals, and 24 concepts. Main entries are unstandardized regression coefficients. Entries in parentheses are standard errors. Entries in italics are standardized regression coefficients. For "Recall score" a Cumulative Link Model was employed: the estimated threshold coefficients are (0|1 estimate = −1.61 (SE = 0.32); 1|2 estimate = −0.29 (SE = 0.31)). The control condition (i.e., no image) was used as the baseline. For image fixations, conditions are compared based on fixations toward the area where the image appeared in other conditions.</item> <item>4 * <emph>p</emph> < 0.05.</item> <item>5 ** <emph>p</emph> < 0.01.</item> <item>6 *** <emph>p <</emph> 0.001.</item> <item>4 TABLE Pairwise comparisons between image‐containing conditions, controlling for familiarity.</item> </ulist> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left" /><th align="center" /><th align="center">Tangentially related image</th><th align="center">Unrelated image</th></tr></thead><tbody valign="top"><tr><td align="left">Keyword fixation duration</td><td align="center">Direct depiction</td><td align="center">−279.03<xref ref-type="fn" rid="tfn8" />(117.89)−0.09</td><td align="center">−523.88<xref ref-type="fn" rid="tfn10" />(117.56)−0.17</td></tr><tr><td align="center">Tangentially related image</td><td align="center" /><td align="center">−244.85<xref ref-type="fn" rid="tfn8" />(118.41)−0.08</td></tr><tr><td align="left">Sentence fixation duration</td><td align="center">Direct depiction</td><td align="center">−738.22<xref ref-type="fn" rid="tfn8" />(333.92)−0.07</td><td align="center">−1539.43<xref ref-type="fn" rid="tfn10" />(332.92)−0.15</td></tr><tr><td align="center">Tangentially related image</td><td align="center" /><td align="center">−801.21<xref ref-type="fn" rid="tfn8" />(335.36)−0.08</td></tr><tr><td align="left">Trial duration</td><td align="center">Direct depiction</td><td align="center">−402.58(435.23)−0.03</td><td align="center">−1425.43<xref ref-type="fn" rid="tfn9" />(433.91)−0.10</td></tr><tr><td align="center">Tangentially related image</td><td align="center" /><td align="center">−1022.85<xref ref-type="fn" rid="tfn8" />(437.11)−0.07</td></tr><tr><td align="left">Recall score</td><td align="center">Direct depiction</td><td align="center">0.16(0.28)0.07</td><td align="center">1.21<xref ref-type="fn" rid="tfn10" />(0.26)0.53</td></tr><tr><td align="center">Tangentially related image</td><td align="center" /><td align="center">1.05<xref ref-type="fn" rid="tfn10" />(0.26)0.45</td></tr><tr><td align="left">Image fixation duration</td><td align="center">Direct depiction</td><td align="center">454.83<xref ref-type="fn" rid="tfn10" />(114.36)0.16</td><td align="center">457.41<xref ref-type="fn" rid="tfn10" />(114.01)0.16</td></tr><tr><td align="center">Tangentially related image</td><td align="center" /><td align="center">2.58(114.88)0.00</td></tr></tbody></table> </ephtml> </p> <ulist> <item>7 <emph>Note:n</emph> = 652 observations, 31 individuals, and 24 keywords. Main entries are unstandardized regression coefficients. Entries in parentheses are standard errors. Entries in italics are standardized regression coefficients. For "Recall score", a Cumulative Link Model was employed. Presented estimates are controlled for familiarity.</item> <item>8 * <emph>p</emph> < 0.05.</item> <item>9 ** <emph>p</emph> < 0.01.</item> <item>10 *** <emph>p <</emph> 0.001.</item> </ulist> <hd id="AN0186111736-22">Familiarity</hd> <p>In all models, the main effect of familiarity significantly enhanced model fit (all <emph>p</emph>s < 0.001). The familiarity effects (<emph>p</emph>s < 0.001) were medium for keywords (<emph>t</emph>(<reflink idref="bib484" id="ref83">484</reflink>) = −7.57), small for sentences (<emph>t</emph>(<reflink idref="bib622" id="ref84">622</reflink>) = −7.04), small for trials (<emph>t</emph>(<reflink idref="bib611" id="ref85">611</reflink>) = −8.09), and medium for recall (<emph>t</emph>(<reflink idref="bib644" id="ref86">644</reflink>) = 3.15) (Table 3). The effects of familiarity on keyword and sentence fixation durations are consistent with studies by Lahoud et al. ([<reflink idref="bib23" id="ref87">23</reflink>]) and Williams and Morris ([<reflink idref="bib63" id="ref88">63</reflink>]), providing evidence for criterion validity, where trials measure processing as expected.</p> <p>As for the interaction effects, neither the quadratic term for familiarity nor their interaction effects with the condition variables were significant (all <emph>p</emph>s > 0.05). This means that the effects of the conditions did not vary significantly alongside familiarity. The reported models thus did not include such interactions.</p> <hd id="AN0186111736-23">Keyword Fixation Durations</hd> <p>The results of the analyses revealed that the effect of the conditions on total keyword fixation durations was significant, <emph>F</emph>(<reflink idref="bib3" id="ref89">3</reflink>, 644) = 11.08, <emph>p</emph> < 0.001. The analyses indicated that total fixation durations on keywords were significantly shorter for the direct depiction condition compared to the tangentially related image condition (<emph>p</emph> = 0.018), the unrelated image condition (<emph>p</emph> < 0.001), and the no‐image condition (<emph>p</emph> < 0.001). The tangentially related image condition had significantly shorter total fixation durations on keywords than the unrelated image condition (<emph>p</emph> = 0.039). For the no‐image condition, the total fixation durations on keywords were significantly longer than those in the direct depiction and the tangentially related image condition (<emph>p</emph> = 0.011). However, the no‐image condition did not significantly differ from the unrelated image condition in total fixation durations on keywords (<emph>p</emph> = 0.631). The standardized regression coefficients in Tables 3 and 4 represent the effect sizes. For keyword fixation durations, the effects of the conditions were small.</p> <hd id="AN0186111736-24">Sentence Fixation Durations</hd> <p>The effect of the conditions on sentence fixation durations was significant, <emph>F</emph>(<reflink idref="bib3" id="ref90">3</reflink>, 644) = 9.06, <emph>p</emph> < 0.001. The analyses revealed that total fixation durations on sentences were significantly shorter for the direct depiction condition compared to the tangentially related image condition (<emph>p</emph> = 0.027), the unrelated image condition (<emph>p</emph> < 0.001), and the no‐image condition (<emph>p</emph> < 0.001). The tangentially related image condition had significantly shorter average fixation durations on sentences compared to the unrelated image condition (<emph>p</emph> = 0.017). The tangentially and unrelated image conditions did not significantly differ from the no‐image condition in total fixation durations on images (<emph>p</emph> = 0.090 and <emph>p</emph> = 0.476, respectively). All significant condition effects on total sentence fixation durations were small.</p> <hd id="AN0186111736-25">Trial Durations</hd> <p>The effect of the conditions on trial durations was significant, <emph>F</emph>(<reflink idref="bib3" id="ref91">3</reflink>, 644) = 5.94, <emph>p</emph> < 0.001. The analysis indicated that trial durations were not significantly different between the direct depiction condition and the tangentially related image condition (<emph>p</emph> = 0.355) or no‐image condition (<emph>p</emph> = 0.687). However, trial durations were significantly shorter in the direct depiction condition compared to the unrelated image condition (<emph>p</emph> = 0.001). Additionally, trial durations were significantly shorter in the tangentially related image condition compared to the unrelated image condition (<emph>p</emph> = 0.020), but not compared to the no‐image condition (<emph>p</emph> = 0.183). Lastly, trial durations were significantly shorter in the no‐image condition compared to the unrelated image condition (<emph>p</emph> < 0.001). All condition effects on trial durations were small.</p> <hd id="AN0186111736-26">Recall Test Scores</hd> <p>The analysis of recall scores did not reveal differences between the direct depiction condition and the tangentially related image condition (<emph>p</emph> = 0.574). However, the analysis indicated that higher recall scores were significantly more likely to occur when concepts were accompanied by direct depictions compared to unrelated images (<emph>p</emph> < 0.001) or no images (<emph>p</emph> < 0.008), with large and moderate differences, respectively. Similarly, higher recall scores were significantly more likely to occur when concepts were accompanied by tangentially related images compared to unrelated images (<emph>p</emph> < 0.001) or no images (<emph>p</emph> < 0.038), with moderate differences. Lastly, compared to the unrelated image condition, the no‐image condition was significantly more likely to yield higher recall scores (<emph>p</emph> = 0.034), with a moderate effect size. All condition effects are visualized in Figure 2.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01may25/acp70054-fig-0002.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70054-fig-0002.jpg" title="2 Standardized means per condition." /> </p> <p></p> <hd id="AN0186111736-28">Image Fixations</hd> <p>Regarding image fixations, results of the multilevel analysis indicated that participants' total time spent fixating on the images differed between the three image‐containing conditions, <emph>F</emph>(<reflink idref="bib2" id="ref92">2</reflink>, 478) = 10.43, <emph>p</emph> < 0.001. Participants spent significantly more time fixating on direct depictions versus less‐related images and unrelated images (<emph>p</emph> < 0.001), with small effects. Participants in the tangentially related and unrelated image conditions did not differ significantly in fixation durations on images (<emph>p</emph> = 0.982).</p> <hd id="AN0186111736-29">Residual Variances</hd> <p>None of the models demonstrated a significant improvement in fit when allowing condition effects to vary between concepts (i.e., random slopes). The reductions in chi‐squared values between models with and without condition parameter variance were all small and insignificant (all <emph>p</emph>s > 0.05). This indicates that there were no discernible differences in the effects of conditions across the different concepts. In other words, the conditions appeared to exert a similar influence regardless of the specific concept being learned.</p> <hd id="AN0186111736-30">Discussion</hd> <p>The present study explored how image relatedness may nuance the coherence principle. Specifically, it examined how images with varying degrees of relatedness to the to‐be‐learned concepts—directly related, tangentially related, and unrelated—affect processing and recall performance. In doing so, it explored whether images, even when not directly relevant for the instructional objective, could still have instructional value by activating semantically related concepts, despite the known potential for such images to induce unnecessary cognitive load (Mayer [<reflink idref="bib31" id="ref93">31</reflink>]).</p> <hd id="AN0186111736-31">Current Findings</hd> <p>The findings provide evidence supporting the notion that images, even when only tangentially related, have the potential to outweigh their extraneous cognitive load. To explore the interplay between this cognitive load and the potential for images to activate semantically linked mental representations, participants' processing and recall of various concepts were assessed. The processing of these concepts, which differed in their relatedness to the displayed images, was evaluated using four measures: fixation duration on keywords and their accompanying sentences, processing time, and recall scores of definitions. The findings illustrate how visual images, when thoughtfully integrated, can contribute positively to the learning process, thereby challenging the prevailing inclination toward the uniform exclusion of visuals that are deemed less relevant.</p> <hd id="AN0186111736-32">Effects of Images on Processing and Recall</hd> <p>The present study found that participants were better able to recall concept definitions when the instructional material paired a concept with a (tangentially) related image. Moreover, participants were more likely to correctly recall definitions when the accompanying images were more closely related to the concept being learned. This suggests that the inclusion of related images enhances learners' ability to recall definitions, likely by facilitating concepts' integration within existing mental frameworks, as postulated by Anderson ([<reflink idref="bib3" id="ref94">3</reflink>]), thereby enhancing the ability to recall their definitions. However, when images were unrelated to the to‐be‐learned concept, impaired recall performance was observed, with unrelated images leading to impaired recall scores compared to no images. Such coherence effects have been commonly observed in previous research (e.g., Rop et al. [<reflink idref="bib45" id="ref95">45</reflink>]; Scharinger [<reflink idref="bib46" id="ref96">46</reflink>]).</p> <p>Similar trends were observed in how participants processed the presented concept. Compared to no images, concepts paired with images that were at least tangentially related led to reduced fixation durations on the presented texts (i.e., keyword and complementary sentences). In general, more related images resulted in greater reductions in processing durations. This trend disappeared for trial durations, which also considered the time spent looking at the images, which showed no differences between related images and no images.</p> <p>As for the negative impact of the images, only for trial duration did the inclusion of an unrelated image increase the overall processing duration compared to more related images or no images. The increased processing demands in trials with an unrelated image were thus higher compared to trials with more related images or no images, which aligns with previous eye‐tracking research (e.g., Scharinger et al. [<reflink idref="bib49" id="ref97">49</reflink>]; Scharinger [<reflink idref="bib46" id="ref98">46</reflink>]). Since such coherence effects did not appear in the text fixation measures, this can only be attributed to the time spent fixating on the image.</p> <p>The image fixation data indicated that the observed relatedness effects cannot be solely attributed to variations in attention to the image: there were no differences in the duration of fixations on tangentially related versus unrelated images, despite the tangentially related image condition outperforming in processing and recall of the concepts. Consequently, there is no evidence to suggest that the relatedness effect is driven by unrelated images being more distracting.</p> <p>Furthermore, the positive effects of related images cannot be solely attributed to reduced schema interference (Rey [<reflink idref="bib42" id="ref99">42</reflink>]), as tangentially related images should theoretically induce a certain degree of interference compared to no images, yet they performed better than the no‐image condition. Therefore, it can be concluded that the observed differences might indeed be attributed to the visual images activating semantically related concepts in memory, thereby facilitating the processing of the definitions to be learned. This image‐relatedness effect not only confirms findings by Scherer et al. ([<reflink idref="bib51" id="ref100">51</reflink>]), replicating their positive effect of related visual images on recall, but also builds upon their findings by showcasing how the benefits vary alongside relatedness to the instructional content.</p> <p>The data also revealed a notable pattern where participants fixated shorter on tangentially and unrelated images compared to direct depictions, which conforms to prior findings regarding learners' ability to discern and circumvent seductive details (Rop et al. [<reflink idref="bib45" id="ref101">45</reflink>]). However, similar image fixation durations between the tangentially and unrelated conditions suggest that participants employed the same avoidance strategies. This implies that the positive effects of tangentially related images mostly occurred implicitly (e.g., Rissman et al. [<reflink idref="bib44" id="ref102">44</reflink>]).</p> <hd id="AN0186111736-33">Coherence Effects Differ per Outcome Variable</hd> <p>The results of the present study offer a comprehensive view on the coherence principle, showing how conclusions and interpretations depend on the specific outcome measure being examined. In terms of how quickly participants processed the material, coherence effects were observed only in the total duration of the trial, where no image resulted in faster processing compared to an unrelated image. However, this coherence effect was not found for fixation durations on the concepts or their definitions. This suggests that, while unrelated images slow down processing, they primarily do so because learners spend time processing the images, rather than because the image's extraneous cognitive load slows down processing of other information. If the extraneous cognitive load imposed by the image had been the primary cause, coherence effects would likely have manifested in the fixation durations on the concepts and sentences as well.</p> <p>Furthermore, the results suggest that participants were less able to recall concepts when they were accompanied by unrelated images. The idea that participants spent time processing the images is not a valid explanation for this coherence effect, as the experiment did not allocate a specific duration for the trial. Therefore, participants' time spent processing the images would not have deprived them of the time needed to process the concept definitions. Cognitive explanations for the coherence effect, such as the images consuming working memory capacity (e.g., Mayer [<reflink idref="bib31" id="ref103">31</reflink>]) or causing schema interference (Rey [<reflink idref="bib42" id="ref104">42</reflink>]), are more likely to explain the observed coherence effect in recall scores.</p> <hd id="AN0186111736-34">Effects of Familiarity With Concepts</hd> <p>Familiarity was measured to control for and test the idea that priming effects would be most profound when concepts were slightly familiar (Ratcliff and McKoon [<reflink idref="bib39" id="ref105">39</reflink>]) but not completely known. However, modeling quadratic interactions between familiarity and the conditions yielded no significant estimates. It remains unclear whether this is due to a lack of power or an actual absence in the population. While potential confounding of such interaction effects did not prevent main effects from being found, the study was unable to explore how they vary with respondents' familiarity. This leaves an interesting avenue for future research.</p> <hd id="AN0186111736-35">Limitations and Future Directions</hd> <p>A limitation of the study was the recall test's limited discriminative capacity. The test was constrained by the concepts' definitions, for which the preference for varied familiarity levels constrained the test to be easier overall. Additionally, many participants read concepts multiple times, resulting in duration outliers that required removal. Future studies exploring image relatedness could thus benefit from implementing a maximum duration.</p> <p>Another limitation of the present study is that cognitive load, while a central theoretical construct, was not measured directly. It was decided not to include this measure as asking for cognitive load estimates 24 times during the experiment (i.e., after the presentation of each concept) would add a large distraction for participants and would compromise practical feasibility. However, the cognitive load induced by images in instructional texts has been extensively studied using physiological measures like fMRI, Electroencephalography, or pupil dilation (e.g., Heath et al. [<reflink idref="bib15" id="ref106">15</reflink>]; Scharinger [<reflink idref="bib47" id="ref107">47</reflink>], [<reflink idref="bib48" id="ref108">48</reflink>]) or self‐report measures (e.g., Park et al. [<reflink idref="bib38" id="ref109">38</reflink>]). Given this extensive body of research, the images in the present study were assumed to induce a significant degree of extraneous cognitive load, and efforts replicating these established effects were not undertaken.</p> <p>A final limitation is the familiarity measure at the end of the experiment. Asking for familiarity after the trials meant that it could have been influenced by the experimental manipulation: the priming may have occurred implicitly (Rissman et al. [<reflink idref="bib44" id="ref110">44</reflink>]), possibly inflating the familiarity measures. However, as one of the outcome variables was the duration respondents fixated on concepts, asking respondents about their familiarity prior to the experiment would have biased the results even more, as it would require respondents to be exposed to the concepts prior to the trial. Asking for familiarity post‐experiment was thus the preferred method.</p> <hd id="AN0186111736-36">Implications</hd> <p></p> <hd id="AN0186111736-37">Theoretical Implications</hd> <p>The results of the present study show that semantic relationships can help explain some of the discrepancies regarding the impact of irrelevant images in instructional texts (Rey [<reflink idref="bib42" id="ref111">42</reflink>]; Sundararajan and Adesope [<reflink idref="bib53" id="ref112">53</reflink>]; Towler et al. [<reflink idref="bib58" id="ref113">58</reflink>]). Furthermore, since seductive details inherently have some relation to the instructional content (e.g., Bender et al. [<reflink idref="bib5" id="ref114">5</reflink>]), it is important to consider that this characteristic may also influence their impact. In this context, the present study highlights how future research on seductive details might benefit from controlling for semantic relations.</p> <hd id="AN0186111736-38">Practical Implications</hd> <p>The findings also offer educators practical insights into utilizing related images to optimize educational materials. In general, educational practitioners can leverage the insights from the present study to make more informed decisions about whether or not to include specific images in their materials. This represents an additional factor to consider when deciding whether to include an image, alongside other benefits of images such as mood enhancement, reduced perceived difficulty, or increased engagement (Alarcão and Fonseca [<reflink idref="bib1" id="ref115">1</reflink>]; Carney and Levin [<reflink idref="bib9" id="ref116">9</reflink>]; Lenzner et al. [<reflink idref="bib25" id="ref117">25</reflink>]; Schneider et al. [<reflink idref="bib52" id="ref118">52</reflink>]).</p> <p>The proposed benefits of somewhat related images are particularly beneficial for abstract concepts that are difficult to represent directly (Borghi et al. [<reflink idref="bib7" id="ref119">7</reflink>]). Moreover, the findings suggest the presence of a critical threshold in relatedness beyond which images transition from being detrimental to learning to facilitating it. This delineates an intriguing area for further investigation, wherein the continuum of relatedness between images and concepts could be quantified further to reflect the varied connectivity between concepts stored in memory (Tulving and Schacter [<reflink idref="bib59" id="ref120">59</reflink>]).</p> <p>However, in applying these findings, it is important to approach the use of related images with caution. Contemporary research suggests that the relatedness of an image could make seductive details even more detrimental, particularly because it conceals their irrelevance (Eitel et al. [<reflink idref="bib12" id="ref121">12</reflink>]). It is difficult to assess the extent to which the current study accounted for this effect, as the experimental design may have emphasized the irrelevant nature of certain images used in the study.</p> <hd id="AN0186111736-39">Transferability</hd> <p>As for the transferability of the findings, it is important to note that participants in the present study were exposed to the images during individual trials. However, in real‐world contexts, such as textbooks, an image often appears alongside an entire page of instructional content that includes multiple meaningful units of information. In such contexts, the image might only be semantically related to specific portions of the instructional content rather than all of it, while still inducing cognitive load during the learning of unrelated information.</p> <p>This interplay is reflected in the findings of the current study, which showed that the priming effects of related images were more pronounced for specific concepts than for whole sentences. This suggests that the more varied the content of the learning material the image is presented alongside, the less impactful the priming benefit of related images might become. This would mean that the positive effects of the images observed in the present study might be stronger compared to images integrated in traditional textbook pages.</p> <hd id="AN0186111736-40">Conclusion</hd> <p>In conclusion, the study highlights that related decorative images can potentially enhance concept processing and recall by eliciting semantically linked mental representations, outweighing any extraneous cognitive load. Related images led to faster processing and better recall capabilities, and coherence effects only appeared when images were completely unrelated. These findings contribute to a better understanding of the conditions under which the coherence principle holds, aiding educators in making informed decisions about the effectiveness of visual aids in instructional materials.</p> <hd id="AN0186111736-41">Author Contributions</hd> <p> <bold>Tristan H. S. de Jonge:</bold> methodology, conceptualization, investigation, writing – original draft, visualization, writing – review and editing, project administration, software, formal analysis, data curation. <bold>Anna Berti:</bold> conceptualization, investigation, writing – original draft, writing – review and editing, methodology, formal analysis, software, visualization. <bold>Sanne van Schijndel:</bold> project administration, software, methodology, writing – review and editing, writing – original draft, investigation, conceptualization, formal analysis. <bold>Margot van Wermeskerken:</bold> supervision, writing – review and editing, conceptualization. <bold>Ellen Kok:</bold> supervision, project administration, formal analysis, software, methodology, conceptualization, writing – review and editing, validation.</p> <hd id="AN0186111736-42">Ethics Statement</hd> <p>This study was approved by the Utrecht University's Faculty Ethical Review Board.</p> <hd id="AN0186111736-43">Consent</hd> <p>Informed consent was obtained from all participants included in this study.</p> <hd id="AN0186111736-44">Conflicts of Interest</hd> <p>The authors declare no conflicts of interest.</p> <hd id="AN0186111736-45">Data Availability Statement</hd> <p>The data supporting the findings of this study are available at Open Science Framework. https://osf.io/evzsw/?view_only=527ead2d91c0414192f25131199fa519.</p> <hd id="AN0186111736-46">A Appendix Concept Learning Task</hd> <p>A1 TABLE Picture conditions per keyword.</p> <p> <ephtml> <table><thead valign="bottom"><tr><th align="left">Keyword</th><th align="center">Picture</th></tr><tr><th align="center">Direct depiction</th><th align="center">Tangentially related</th><th align="center">Non‐related</th></tr></thead><tbody valign="top"><tr><td align="left">Ferrule</td><td align="center">Picture of the metal tip of a pencil with an eraser attached.</td><td align="center">Picture of a pen.</td><td align="center">Picture of an orange bouncing ball.</td></tr><tr><td align="left">Glass ceiling</td><td align="center">Picture a woman positioned amidst exclusively male leaders, all donned in impressive uniforms.</td><td align="center">Picture of high‐level leadership individuals casting votes in a room.</td><td align="center">Picture of a man splitting a piece of wood in half.</td></tr><tr><td align="left">Concussion</td><td align="center">Picture of a girl with an ice pack on her head.</td><td align="center">Picture of two football players tripping over each other.</td><td align="center">Picture of two people cooking.</td></tr><tr><td align="left">Insulin</td><td align="center">Picture of a stomach with an insulin pump attached.</td><td align="center">Picture of a row of beds in the hospital.</td><td align="center">Picture of a dishwasher.</td></tr><tr><td align="left">Meningitis</td><td align="center">Picture of a skeleton, with reference to the brain where you can see a membrane infection.</td><td align="center">Picture of someone with a stethoscope in hand.</td><td align="center">Picture of a basil plant.</td></tr><tr><td align="left">Morphine</td><td align="center">Picture of a syringe with a tube of liquid and a pill.</td><td align="center">Picture of a girl smoking a cannabis cigarette.</td><td align="center">Picture of a starfish.</td></tr><tr><td align="left">Zone of Proximal Development</td><td align="center">Picture of a father helping his son learn to ride a bike.</td><td align="center">Picture of a family walking together in the forest.</td><td align="center">Picture of a woman standing in front of a field of tulips.</td></tr><tr><td align="left">Formative assessment</td><td align="center">Picture of a sheet of paper with comments made in red pen.</td><td align="center">Picture of a classroom.</td><td align="center">Picture of a tree.</td></tr><tr><td align="left">Glabella</td><td align="center">Picture of the skin part between two eyebrows.</td><td align="center">Picture of a mouth surrounded by a brown beard.</td><td align="center">Picture of glass bottles of different colors.</td></tr><tr><td align="left">Delayed gratification</td><td align="center">Picture of a boy looking at a marshmallow.</td><td align="center">Picture of three children playing with building blocks.</td><td align="center">Picture of a man tying his shoelaces.</td></tr><tr><td align="left">Neurotransmitter</td><td align="center">Picture of a substance being passed between two brain cells.</td><td align="center">Picture of a cross‐section of the torso of the human body.</td><td align="center">Picture of a sword.</td></tr><tr><td align="left">Distillation</td><td align="center">Picture of a distillation.</td><td align="center">Picture of a pan with boiling water.</td><td align="center">Picture of a city.</td></tr><tr><td align="left">Eclipse</td><td align="center">Picture of solar eclipse.</td><td align="center">Picture of a rocket taking off.</td><td align="center">Picture of carnivorous plants.</td></tr><tr><td align="left">Paleontology</td><td align="center">Picture of a skeleton of a prehistoric animal on the ground.</td><td align="center">Picture of a prehistoric book.</td><td align="center">Picture of a slot machine.</td></tr><tr><td align="left">Erosion</td><td align="center">Picture of cracks in the ground.</td><td align="center">Picture of snow at the top of a mountain.</td><td align="center">Picture of a pencil sharpened by a sharpener.</td></tr><tr><td align="left">Augmented reality</td><td align="center">Picture of an iPad displaying the city.</td><td align="center">Picture of a charger.</td><td align="center">Picture of two binoculars.</td></tr><tr><td align="left">Jackal</td><td align="center">Picture of a jackal.</td><td align="center">Picture of a box of dog biscuits.</td><td align="center">Picture of a glass of juice.</td></tr><tr><td align="left">Yucca</td><td align="center">Picture of a yucca.</td><td align="center">Picture of a watering can.</td><td align="center">Picture of a paint set.</td></tr><tr><td align="left">Trebuchet</td><td align="center">Picture of a trebuchet.</td><td align="center">Picture of a medieval suit of armor.</td><td align="center">Picture of a lighter.</td></tr><tr><td align="left">Bifocal</td><td align="center">Picture of glasses.</td><td align="center">Picture of a hearing aid.</td><td align="center">Picture of trees.</td></tr><tr><td align="left">Amphibious vehicle</td><td align="center">Picture of an amphibious vehicle.</td><td align="center">Picture of tires from a military vehicle.</td><td align="center">Picture of a hole puncher in a belt.</td></tr><tr><td align="left">Siphon</td><td align="center">Picture of a siphon.</td><td align="center">Picture of a toilet plunger.</td><td align="center">Picture of a pomegranate.</td></tr><tr><td align="left">Pergola</td><td align="center">Picture of a pergola.</td><td align="center">Picture of a garden.</td><td align="center">Picture of a gaming console.</td></tr><tr><td align="left">Soil profile</td><td align="center">Picture of a soil profile (wall of a hole in the ground in which different layers of soil can be distinguished).</td><td align="center">Picture of someone digging a hole in the sand.</td><td align="center">Picture of a seagull.</td></tr></tbody></table> </ephtml> </p> <ref id="AN0186111736-47"> <title> Footnotes </title> <blist> <bibl id="bib1" idref="ref55" type="bt">1</bibl> <bibtext> Link to dataset: https://osf.io/evzsw/?view_only=dccf5056ca31476b8125dfb67059a9e0.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref44" type="bt">2</bibl> <bibtext> Funding: The authors received no specific funding for this work.</bibtext> </blist> </ref> <ref id="AN0186111736-48"> <title> References </title> <blist> <bibtext> Alarcão, S. M., and M. J. Fonseca. 2017. " Identifying Emotions in Images From Valence and Arousal Ratings." Multimedia Tools and Applications 77, no. 13 : 17413 – 17435.</bibtext> </blist> <blist> <bibtext> Anaki, D., and A. Henik. 2003. " Is There a "Strength Effect" in Automatic Semantic Priming? " Memory & Cognition 31, no. 2 : 262 – 272. https://doi.org/10.3758/bf03194385.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref33" type="bt">3</bibl> <bibtext> Anderson, J. R. 1983. " A Spreading Activation Theory of Memory." Journal of Verbal Learning and Verbal Behavior 22, no. 3 : 261 – 295. https://doi.org/10.1016/S0022‐5371(83)90201‐3.</bibtext> </blist> <blist> <bibl id="bib4" idref="ref4" type="bt">4</bibl> <bibtext> Angwin, A. J., H. J. Chenery, D. A. Copland, W. L. Arnott, B. E. Murdoch, and P. A. Silburn. 2004. " Dopamine and Semantic Activation: An Investigation of Masked Direct and Indirect Priming." Journal of the International Neuropsychological Society 10, no. 1 : 15 – 25. https://doi.org/10.1017/s1355617704101033.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref21" type="bt">5</bibl> <bibtext> Bender, L., A. Renkl, and A. Eitel. 2021. " Seductive Details Do Their Damage Also in Longer Learning Sessions – When the Details Are Perceived as Relevant." Journal of Computer Assisted Learning 37, no. 5 : 1248 – 1262. https://doi.org/10.1111/jcal.12560.</bibtext> </blist> <blist> <bibl id="bib6" idref="ref69" type="bt">6</bibl> <bibtext> Beymer, D., P. Z. Orton, and D. M. Russell. 2007. " An Eye Tracking Study of How Pictures Influence Online Reading." In Human‐Computer Interaction – INTERACT 2007, edited by C. Baranauskas, P. Palanque, J. Abascal, and S. D. J. Barbosa, 456 – 460. Springer. https://doi.org/10.1007/978‐3‐540‐74800‐7_41.</bibtext> </blist> <blist> <bibl id="bib7" idref="ref119" type="bt">7</bibl> <bibtext> Borghi, A. M., F. Binkofski, C. Castelfranchi, F. Cimatti, C. Scorolli, and L. Tummolini. 2017. " The Challenge of Abstract Concepts." Psychological Bulletin 143, no. 3 : 263 – 292. https://doi.org/10.1037/bul0000089.</bibtext> </blist> <blist> <bibl id="bib8" idref="ref79" type="bt">8</bibl> <bibtext> Brysbaert, M. 2019. " How Many Words Do We Read per Minute? A Review and Meta‐Analysis of Reading Rate." Journal of Memory and Language 109 : 104047. https://doi.org/10.1016/j.jml.2019.104047.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref16" type="bt">9</bibl> <bibtext> Carney, R. N., and J. R. Levin. 2002. " Pictorial Illustrations Still Improve Students' Learning From Text." Educational Psychology Review 14, no. 1 : 5 – 26. https://doi.org/10.1023/a:1013176309260.</bibtext> </blist> <blist> <bibtext> Christensen, R. H. B. 2015. "Analysis of Ordinal Data With Cumulative Link Models —Estimation With the R‐Package Ordinal." R‐Package Version, 28, 406. https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=ebbe4c4abe64c3889bc3309311790c367fd9bf51.</bibtext> </blist> <blist> <bibtext> Cohen, J. 1988. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Lawrence Erlbaum Associates.</bibtext> </blist> <blist> <bibtext> Eitel, A., L. Bender, and A. Renkl. 2018. " Are Seductive Details Seductive Only When You Think They Are Relevant? An Experimental Test of the Moderating Role of Perceived Relevance." Applied Cognitive Psychology 33, no. 1 : 20 – 30. https://doi.org/10.1002/acp.3479.</bibtext> </blist> <blist> <bibtext> Elamir, E. A., and A. H. Seheult. 2003. " Trimmed L‐Moments." Computational Statistics & Data Analysis 43, no. 3 : 299 – 314. https://doi.org/10.1016/s0167‐9473(02)00250‐5.</bibtext> </blist> <blist> <bibtext> Garner, R., M. G. Gillingham, and C. S. White. 1989. " Effects of 'Seductive Details' on Macroprocessing and Microprocessing in Adults and Children." Cognition and Instruction 6, no. 1 : 41 – 57. https://doi.org/10.1207/s1532690xci0601_2.</bibtext> </blist> <blist> <bibtext> Heath, S., K. McMahon, L. Nickels, et al. 2012. " Priming Picture Naming With a Semantic Task: An fMRI Investigation." PLoS One 7, no. 3 : e32809. https://doi.org/10.1371/journal.pone.0032809.</bibtext> </blist> <blist> <bibtext> Hedeker, D. 2008. Handbook of Multilevel Analysis, 237 – 274. Springer New York. https://doi.org/10.1007/978‐0‐387‐73186‐5_6.</bibtext> </blist> <blist> <bibtext> Hox, J. J., M. Moerbeek, and R. Van De Schoot. 2017. Multilevel Analysis. Routledge eBooks. https://doi.org/10.4324/9781315650982.</bibtext> </blist> <blist> <bibtext> Hu, L., G. Chen, P. Li, and J. Huang. 2021. " Multimedia Effect in Problem Solving: A Meta‐Analysis." Educational Psychology Review 33, no. 4 : 1717 – 1747. https://doi.org/10.1007/s10648‐021‐09610‐z.</bibtext> </blist> <blist> <bibtext> Kim, H. 2016. " Brain Regions That Show Repetition Suppression and Enhancement: A Meta‐Analysis of 137 Neuroimaging Experiments." Human Brain Mapping 38, no. 4 : 1894 – 1913. https://doi.org/10.1002/hbm.23492.</bibtext> </blist> <blist> <bibtext> Kircher, T., K. Sass, O. Sachs, and S. Krach. 2009. " Priming Words With Pictures: Neural Correlates of Semantic Associations in a Cross‐Modal Priming Task Using fMRI." Human Brain Mapping 30, no. 12 : 4116 – 4128. https://doi.org/10.1002/hbm.20833.</bibtext> </blist> <blist> <bibtext> Korbach, A., R. Brünken, and B. Park. 2016. " Learner Characteristics and Information Processing in Multimedia Learning: A Moderated Mediation of the Seductive Details Effect." Learning and Individual Differences 51 : 59 – 68. https://doi.org/10.1016/j.lindif.2016.08.030.</bibtext> </blist> <blist> <bibtext> Korbach, A., R. Brünken, and B. Park. 2017. " Measurement of Cognitive Load in Multimedia Learning: A Comparison of Different Objective Measures." Instructional Science 45 : 515 – 536. https://doi.org/10.1007/s11251‐017‐9413‐5.</bibtext> </blist> <blist> <bibtext> Lahoud, H., D. L. Share, and A. Shechter. 2023. " A Developmental Study of Eye Movements in Hebrew Word Reading: The Effects of Word Familiarity, Word Length, and Reading Proficiency." Frontiers in Psychology 14 : 1052755. https://doi.org/10.3389/fpsyg.2023.1052755.</bibtext> </blist> <blist> <bibtext> Lehman, S., G. Schraw, M. T. McCrudden, and K. Hartley. 2007. " Processing and Recall of Seductive Details in Scientific Text." Contemporary Educational Psychology 32, no. 4 : 569 – 587. https://doi.org/10.1016/j.cedpsych.2006.07.002.</bibtext> </blist> <blist> <bibtext> Lenzner, A., W. Schnotz, and A. Müller. 2013. " The Role of Decorative Pictures in Learning." Instructional Science 41, no. 5 : 811 – 831. https://doi.org/10.1007/s11251‐012‐9256‐z.</bibtext> </blist> <blist> <bibtext> Levie, W. H., and R. Lentz. 1982. " Effects of Text Illustrations: A Review of Research." Educational Technology Research and Development 30, no. 4 : 195 – 232. https://doi.org/10.1007/bf02765184.</bibtext> </blist> <blist> <bibtext> Levin, J. R., G. J. Anglin, and R. N. Carney. 1987. " On Empirically Validating Functions of Pictures in Prose." In The Psychology of Illustration: I. Basic Research, edited by D. M. Willows and H. A. Houghton, 51 – 85. Springer.</bibtext> </blist> <blist> <bibtext> Lindner, M. A. 2020. " Representational and Decorative Pictures in Science and Mathematics Tests: Do They Make a Difference? " Learning and Instruction 68 : 101345. https://doi.org/10.1016/j.learninstruc.2020.101345.</bibtext> </blist> <blist> <bibtext> Lindner, M. A., J. M. Ihme, S. Saß, and O. Köller. 2018. " How Representational Pictures Enhance Students' Performance and Test‐Taking Pleasure in Low‐Stakes Assessment." European Journal of Psychological Assessment 34, no. 6 : 376 – 385. https://doi.org/10.1027/1015‐5759/a000351.</bibtext> </blist> <blist> <bibtext> Mayer, R. E. 1997. " Multimedia Learning: Are We Asking the Right Questions? " Educational Psychologist 32 : 1 – 19.</bibtext> </blist> <blist> <bibtext> Mayer, R. E. 2005. The Cambridge Handbook of Multimedia Learning. Cambridge university press. https://doi.org/10.1017/CBO9780511816819.</bibtext> </blist> <blist> <bibtext> Mayer, R. E. 2009. Multimedia Learning. 2nd ed. Cambridge University Press.</bibtext> </blist> <blist> <bibtext> McHugh, M. L. 2012. " Interrater Reliability: The Kappa Statistic." Biochemia Medica 22, no. 3 : 276 – 282.</bibtext> </blist> <blist> <bibtext> McNeish, D. M., and L. M. Stapleton. 2014. " The Effect of Small Sample Size on Two‐Level Model Estimates: A Review and Illustration." Educational Psychology Review 28, no. 2 : 295 – 314. https://doi.org/10.1007/s10648‐014‐9287‐x.</bibtext> </blist> <blist> <bibtext> Meyer, D. E., and R. W. Schvaneveldt. 1971. " Facilitation in Recognizing Pairs of Words: Evidence of a Dependence Between Retrieval Operations." Journal of Experimental Psychology 90, no. 2 : 227 – 234. https://doi.org/10.1037/h0031564.</bibtext> </blist> <blist> <bibtext> Moreno, R., and R. E. Mayer. 1999. " Cognitive Principles of Multimedia Learning: The Role of Modality and Contiguity." Journal of Educational Psychology 91, no. 2 : 358 – 368. https://doi.org/10.1037/0022‐0663.91.2.358.</bibtext> </blist> <blist> <bibtext> Neely, J. H. 1977. " Semantic Priming and Retrieval From Lexical Memory: Roles of Inhibitionless Spreading Activation and Limited‐Capacity Attention." Journal of Experimental Psychology: General 106, no. 3 : 226 – 254. https://doi.org/10.1037/0096‐3445.106.3.226.</bibtext> </blist> <blist> <bibtext> Park, B., R. Moreno, T. Seufert, and R. Brünken. 2011. " Does Cognitive Load Moderate the Seductive Details Effect? A Multimedia Study." Computers in Human Behavior 27, no. 1 : 5 – 10. https://doi.org/10.1016/j.chb.2010.05.006.</bibtext> </blist> <blist> <bibtext> Ratcliff, R., and G. McKoon. 1988. " A Retrieval Theory of Priming in Memory." Psychological Review 95, no. 3 : 385 – 408. https://doi.org/10.1037/0033‐295X.95.3.385.</bibtext> </blist> <blist> <bibtext> Rayner, K. 1998. " Eye Movements in Reading and Information Processing: 20 years of Research." Psychological Bulletin 124, no. 3 : 372 – 422. https://doi.org/10.1037/0033‐2909.124.3.372.</bibtext> </blist> <blist> <bibtext> Reichle, E. D., K. Rayner, and A. Pollatsek. 2012. " Eye Movements in Reading Versus Nonreading Tasks: Using EZ Reader to Understand the Role of Word/Stimulus Familiarity." Visual Cognition 20, no. 4–5 : 360 – 390. https://doi.org/10.1080/13506285.2012.667006.</bibtext> </blist> <blist> <bibtext> Rey, G. D. 2012. " A Review of Research and a Meta‐Analysis of the Seductive Detail Effect." Educational Research Review 7, no. 3 : 216 – 237. https://doi.org/10.1016/j.edurev.2012.05.003.</bibtext> </blist> <blist> <bibtext> Rips, L. J., E. J. Shoben, and E. E. Smith. 1973. " Semantic Distance and the Verification of Semantic Relations." Journal of Verbal Learning and Verbal Behavior 12, no. 1 : 1 – 20. https://doi.org/10.1016/s0022‐5371(73)80056‐8.</bibtext> </blist> <blist> <bibtext> Rissman, J., J. C. Eliassen, and S. E. Blumstein. 2003. " An Event‐Related FMRI Investigation of Implicit Semantic Priming." Journal of Cognitive Neuroscience 15, no. 8 : 1160 – 1175. https://doi.org/10.1162/089892903322598120.</bibtext> </blist> <blist> <bibtext> Rop, G., M. Van Wermeskerken, J. A. De Nooijer, P. P. J. L. Verkoeijen, and T. Van Gog. 2016. " Task Experience as a Boundary Condition for the Negative Effects of Irrelevant Information on Learning." Educational Psychology Review 30, no. 1 : 229 – 253. https://doi.org/10.1007/s10648‐016‐9388‐9.</bibtext> </blist> <blist> <bibtext> Scharinger, C. 2022. " Effects of Decorative Pictures on Mental Processing Demands and Learning: An EEG and Eye‐Tracking Study." Acta Psychologica 231 : 103798. https://doi.org/10.1016/j.actpsy.2022.103798.</bibtext> </blist> <blist> <bibtext> Scharinger, C. 2023. " Effects of Emotional Decorative Pictures on Cognitive Load as Assessed by Pupil Dilation and EEG Frequency Band Power." Applied Cognitive Psychology 37, no. 4 : 861 – 875. https://doi.org/10.1002/acp.4087.</bibtext> </blist> <blist> <bibtext> Scharinger, C. 2024. " Task‐Irrelevant Decorative Pictures Increase Cognitive Load During Text Processing but Have no Effects on Learning or Working Memory Performance: An EEG and Eye‐Tracking Study." Psychological Research 88, no. 4 : 1362 – 1388. https://doi.org/10.1007/s00426‐024‐01939‐8.</bibtext> </blist> <blist> <bibtext> Scharinger, C., A. Schüler, and P. Gerjets. 2020. " Using Eye‐Tracking and EEG to Study the Mental Processing Demands During Learning of Text‐Picture Combinations." International Journal of Psychophysiology 158 : 201 – 214. https://doi.org/10.1016/j.ijpsycho.2020.09.014.</bibtext> </blist> <blist> <bibtext> Scheiter, K., P. Gerjets, and E. Heise. 2014. " Distraction During Learning With Hypermedia: Difficult Tasks Help to Keep Task Goals on Track." Frontiers in Psychology 5 : 268. https://doi.org/10.3389/fpsyg.2014.00268.</bibtext> </blist> <blist> <bibtext> Scherer, D., A. Verkühlen, and S. Dutke. 2023. " Effects of Related Decorative Pictures on Learning and Metacognition." Instructional Science 51, no. 4 : 571 – 594. https://doi.org/10.1007/s11251‐023‐09618‐8.</bibtext> </blist> <blist> <bibtext> Schneider, S., S. Nebel, and G. D. Rey. 2016. " Decorative Pictures and Emotional Design in Multimedia Learning." Learning and Instruction 44 : 65 – 73. https://doi.org/10.1016/j.learninstruc.2016.03.002.</bibtext> </blist> <blist> <bibtext> Sundararajan, N., and O. Adesope. 2020. " Keep It Coherent: A Meta‐Analysis of the Seductive Details Effect." Educational Psychology Review 32, no. 3 : 707 – 734. https://doi.org/10.1007/s10648‐020‐09522‐4.</bibtext> </blist> <blist> <bibtext> Sung, E., and R. E. Mayer. 2012. " When Graphics Improve Liking but Not Learning From Online Lessons." Computers in Human Behavior 28, no. 5 : 1618 – 1625. https://doi.org/10.1016/j.chb.2012.03.026.</bibtext> </blist> <blist> <bibtext> Takahashi, S. 1995. " Aesthetic Properties of Pictorial Perception." Psychological Review 102, no. 4 : 671 – 683. https://doi.org/10.1037/0033‐295x.102.4.671.</bibtext> </blist> <blist> <bibtext> Tivarus, M. E., J. W. Ibinson, A. Hillier, P. Schmalbrock, and D. Q. Beversdorf. 2006. " An fMRI Study of Semantic Priming: Modulation of Brain Activity by Varying Semantic Distances." Cognitive and Behavioral Neurology 19, no. 4 : 194 – 201. https://doi.org/10.1097/01.wnn.0000213913.87642.74.</bibtext> </blist> <blist> <bibtext> Tobii. 2023. "Tobii Pro Spectrum: Deeper Insights into Human Behavior." https://<ulink href="http://www.tobii.com/products/eye‐trackers/screen‐based/tobii‐pro‐spectrum">www.tobii.com/products/eye‐trackers/screen‐based/tobii‐pro‐spectrum</ulink>.</bibtext> </blist> <blist> <bibtext> Towler, A., K. Kraiger, T. Sitzmann, et al. 2008. " The Seductive Details Effect in Technology‐Delivered Instruction." Performance Improvement Quarterly 21, no. 2 : 65 – 86. https://doi.org/10.1002/piq.20023.</bibtext> </blist> <blist> <bibtext> Tulving, E., and D. L. Schacter. 1990. " Priming and Human Memory Systems." Science 247, no. 4940 : 301 – 306.</bibtext> </blist> <blist> <bibtext> Wagner, A. D., J. E. Desmond, J. B. Demb, G. H. Glover, and J. D. E. Gabrieli. 1997. " Semantic Repetition Priming for Verbal and Pictorial Knowledge: A Functional MRI Study of Left Inferior Prefrontal Cortex." Journal of Cognitive Neuroscience 9, no. 6 : 714 – 726. https://doi.org/10.1162/jocn.1997.9.6.714.</bibtext> </blist> <blist> <bibtext> Wang, Z., and O. Adesope. 2016. " Does Learners' Prior Knowledge Moderate the Detrimental Effects of Seductive Details in Reading From Text? A 2 by 3 Study." International Journal of Instruction 9, no. 2 : 35 – 50. https://doi.org/10.12973/iji.2016.923a.</bibtext> </blist> <blist> <bibtext> Wang, Z., Y. Ardasheva, K. Carbonneau, and Q. Liu. 2021. " Testing the Seductive Details Effect: Does the Format or the Amount of Seductive Details Matter? " Applied Cognitive Psychology 35, no. 3 : 761 – 774. https://doi.org/10.1002/acp.3801.</bibtext> </blist> <blist> <bibtext> Williams, R., and R. Morris. 2004. " Eye Movements, Word Familiarity, and Vocabulary Acquisition." European Journal of Cognitive Psychology 16, no. 1–2 : 312 – 339. https://doi.org/10.1080/09541440340000196.</bibtext> </blist> </ref> <aug> <p>By Tristan H. S. de Jonge; Anna Berti; Sanne van Schijndel; Margot van Wermeskerken and Ellen Kok</p> <p>Reported by Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib42" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib21" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib53" firstref="ref3"></nolink> <nolink nlid="nl4" bibid="bib25" firstref="ref6"></nolink> <nolink nlid="nl5" bibid="bib18" firstref="ref7"></nolink> <nolink nlid="nl6" bibid="bib26" firstref="ref8"></nolink> <nolink nlid="nl7" bibid="bib27" firstref="ref9"></nolink> <nolink nlid="nl8" bibid="bib28" firstref="ref10"></nolink> <nolink nlid="nl9" bibid="bib29" firstref="ref11"></nolink> <nolink nlid="nl10" bibid="bib32" firstref="ref12"></nolink> <nolink nlid="nl11" bibid="bib30" firstref="ref13"></nolink> <nolink nlid="nl12" bibid="bib36" firstref="ref15"></nolink> <nolink nlid="nl13" bibid="bib55" firstref="ref17"></nolink> <nolink nlid="nl14" bibid="bib14" firstref="ref19"></nolink> <nolink nlid="nl15" bibid="bib31" firstref="ref20"></nolink> <nolink nlid="nl16" bibid="bib54" firstref="ref22"></nolink> <nolink nlid="nl17" bibid="bib24" firstref="ref24"></nolink> <nolink nlid="nl18" bibid="bib22" firstref="ref27"></nolink> <nolink nlid="nl19" bibid="bib50" firstref="ref30"></nolink> <nolink nlid="nl20" bibid="bib61" firstref="ref31"></nolink> <nolink nlid="nl21" bibid="bib62" firstref="ref32"></nolink> <nolink nlid="nl22" bibid="bib37" firstref="ref35"></nolink> <nolink nlid="nl23" bibid="bib35" firstref="ref36"></nolink> <nolink nlid="nl24" bibid="bib59" firstref="ref37"></nolink> <nolink nlid="nl25" bibid="bib15" firstref="ref38"></nolink> <nolink nlid="nl26" bibid="bib56" firstref="ref39"></nolink> <nolink nlid="nl27" bibid="bib20" firstref="ref40"></nolink> <nolink nlid="nl28" bibid="bib60" firstref="ref41"></nolink> <nolink nlid="nl29" bibid="bib19" firstref="ref42"></nolink> <nolink nlid="nl30" bibid="bib39" firstref="ref43"></nolink> <nolink nlid="nl31" bibid="bib58" firstref="ref51"></nolink> <nolink nlid="nl32" bibid="bib51" firstref="ref52"></nolink> <nolink nlid="nl33" bibid="bib52" firstref="ref54"></nolink> <nolink nlid="nl34" bibid="bib49" firstref="ref58"></nolink> <nolink nlid="nl35" bibid="bib40" firstref="ref59"></nolink> <nolink nlid="nl36" bibid="bib34" firstref="ref66"></nolink> <nolink nlid="nl37" bibid="bib17" firstref="ref67"></nolink> <nolink nlid="nl38" bibid="bib43" firstref="ref68"></nolink> <nolink nlid="nl39" bibid="bib10" firstref="ref71"></nolink> <nolink nlid="nl40" bibid="bib23" firstref="ref72"></nolink> <nolink nlid="nl41" bibid="bib41" firstref="ref73"></nolink> <nolink nlid="nl42" bibid="bib63" firstref="ref74"></nolink> <nolink nlid="nl43" bibid="bib33" firstref="ref75"></nolink> <nolink nlid="nl44" bibid="bib57" firstref="ref76"></nolink> <nolink nlid="nl45" bibid="bib16" firstref="ref78"></nolink> <nolink nlid="nl46" bibid="bib13" firstref="ref80"></nolink> <nolink nlid="nl47" bibid="bib11" firstref="ref81"></nolink> <nolink nlid="nl48" bibid="bib484" firstref="ref83"></nolink> <nolink nlid="nl49" bibid="bib622" firstref="ref84"></nolink> <nolink nlid="nl50" bibid="bib611" firstref="ref85"></nolink> <nolink nlid="nl51" bibid="bib644" firstref="ref86"></nolink> <nolink nlid="nl52" bibid="bib45" firstref="ref95"></nolink> <nolink nlid="nl53" bibid="bib46" firstref="ref96"></nolink> <nolink nlid="nl54" bibid="bib44" firstref="ref102"></nolink> <nolink nlid="nl55" bibid="bib47" firstref="ref107"></nolink> <nolink nlid="nl56" bibid="bib48" firstref="ref108"></nolink> <nolink nlid="nl57" bibid="bib38" firstref="ref109"></nolink> <nolink nlid="nl58" bibid="bib12" firstref="ref121"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1474588
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Unraveling Coherence: The Impact of Image-Relatedness in Concept Learning
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Tristan+H%2E+S%2E+de+Jonge%22">Tristan H. S. de Jonge</searchLink> (ORCID <externalLink term="https://orcid.org/0009-0003-9983-4223">0009-0003-9983-4223</externalLink>)<br /><searchLink fieldCode="AR" term="%22Anna+Berti%22">Anna Berti</searchLink> (ORCID <externalLink term="https://orcid.org/0009-0007-2837-6770">0009-0007-2837-6770</externalLink>)<br /><searchLink fieldCode="AR" term="%22Sanne+van+Schijndel%22">Sanne van Schijndel</searchLink><br /><searchLink fieldCode="AR" term="%22Margot+van+Wermeskerken%22">Margot van Wermeskerken</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0003-3883-9875">0000-0003-3883-9875</externalLink>)<br /><searchLink fieldCode="AR" term="%22Ellen+Kok%22">Ellen Kok</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0001-9752-2531">0000-0001-9752-2531</externalLink>)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Applied+Cognitive+Psychology%22"><i>Applied Cognitive Psychology</i></searchLink>. 2025 39(3).
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Wiley. Available from: John Wiley & Sons, Inc. 111 River Street, Hoboken, NJ 07030. Tel: 800-835-6770; e-mail: cs-journals@wiley.com; Web site: https://www.wiley.com/en-us
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 16
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2025
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Concept+Teaching%22">Concept Teaching</searchLink><br /><searchLink fieldCode="DE" term="%22Concept+Formation%22">Concept Formation</searchLink><br /><searchLink fieldCode="DE" term="%22Learning+Processes%22">Learning Processes</searchLink><br /><searchLink fieldCode="DE" term="%22Schemata+%28Cognition%29%22">Schemata (Cognition)</searchLink><br /><searchLink fieldCode="DE" term="%22Imagery%22">Imagery</searchLink><br /><searchLink fieldCode="DE" term="%22Sentences%22">Sentences</searchLink><br /><searchLink fieldCode="DE" term="%22Semantics%22">Semantics</searchLink><br /><searchLink fieldCode="DE" term="%22Priming%22">Priming</searchLink><br /><searchLink fieldCode="DE" term="%22Recall+%28Psychology%29%22">Recall (Psychology)</searchLink><br /><searchLink fieldCode="DE" term="%22Familiarity%22">Familiarity</searchLink><br /><searchLink fieldCode="DE" term="%22Difficulty+Level%22">Difficulty Level</searchLink><br /><searchLink fieldCode="DE" term="%22Instructional+Design%22">Instructional Design</searchLink><br /><searchLink fieldCode="DE" term="%22Instructional+Materials%22">Instructional Materials</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1002/acp.70054
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0888-4080<br />1099-0720
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: The coherence principle suggests removing unnecessary--or seductive--content from educational texts to reduce cognitive load. However, the binary proposition that all seductive details should be excluded neglects images' potential to prime semantically related concepts, which makes texts easier to process. It was hypothesized that this priming would cause at least tangentially related images to enhance processing and recall of concepts. Participants learned 24 concepts under four conditions: direct depictions, tangentially related and unrelated images, and no image. Participants' fixation durations on concepts, their complementing sentences and images, and recall performance were measured. Multilevel models revealed that coherence effects were only present for unrelated images and that images that are at least tangentially related facilitated learning. These effects were unaffected by participants' familiarity with concepts. The study concludes that semantically related images may outweigh their cognitive load, suggesting that educators should consider their priming potential when designing instructional materials.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: Note
  Label: Notes
  Group: Note
  Data: https://osf.io/evzsw/?view_only=527ead2d91c0414192f25131199fa519
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2025
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1474588
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1474588
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1002/acp.70054
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 16
    Subjects:
      – SubjectFull: Concept Teaching
        Type: general
      – SubjectFull: Concept Formation
        Type: general
      – SubjectFull: Learning Processes
        Type: general
      – SubjectFull: Schemata (Cognition)
        Type: general
      – SubjectFull: Imagery
        Type: general
      – SubjectFull: Sentences
        Type: general
      – SubjectFull: Semantics
        Type: general
      – SubjectFull: Priming
        Type: general
      – SubjectFull: Recall (Psychology)
        Type: general
      – SubjectFull: Familiarity
        Type: general
      – SubjectFull: Difficulty Level
        Type: general
      – SubjectFull: Instructional Design
        Type: general
      – SubjectFull: Instructional Materials
        Type: general
    Titles:
      – TitleFull: Unraveling Coherence: The Impact of Image-Relatedness in Concept Learning
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Tristan H. S. de Jonge
      – PersonEntity:
          Name:
            NameFull: Anna Berti
      – PersonEntity:
          Name:
            NameFull: Sanne van Schijndel
      – PersonEntity:
          Name:
            NameFull: Margot van Wermeskerken
      – PersonEntity:
          Name:
            NameFull: Ellen Kok
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 05
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 0888-4080
            – Type: issn-electronic
              Value: 1099-0720
          Numbering:
            – Type: volume
              Value: 39
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Applied Cognitive Psychology
              Type: main
ResultId 1