Two Emotional Design Features Are More Effective Than One in Multimedia Learning

Abstract

This study examined the impacts of adding emotional design features to a multimedia lesson (color alone, anthropomorphism alone, or color & anthropomorphism together) on college students’ affective processes (measured by ratings of experienced emotion during learning), cognitive processes (measured by eye-tracking metrics), and learning outcomes (measured by retention and transfer test scores). One-hundred students were randomly assigned to watch a short multimedia lesson in one of four conditions: no emotional design, colorful emotional design, anthropomorphism emotional design, and colorful and anthropomorphism emotional design. The study results showed that compared to the no emotional design group, the colorful and anthropomorphism emotional design group showed the higher positive emotion rating (d = .726), the shortest time to first fixation on an emotional design area (d = - .877), the longest fixation duration on emotional design areas (d = .640), and the best transfer test score (d = .679). In contrast, the anthropomorphism emotional design group outperformed the no emotional design group only on rating of positive emotion, and the colorful emotional design group outperformed the no emotional design group only on transfer test score. The results show that two emotional design features are more effective than one in multimedia lessons. A structural equation model indicated that positive emotion (tapping affective processing) and fixation duration (tapping cognitive processing) mediated the pathway from emotional design to learning performance. These results partially support the Cognitive-Affective Model of E-Learning.

Keywords

emotional design multimedia learning affective processes cognitive processes eye movements

Introduction

Objective and Rationale

The goal of the present study is to investigate the role of emotional design of the essential elements in a multimedia lesson. In particular, the goal is to determine whether adding two emotional design features has a greater effect than adding one feature on affective processing (as measured by a rating scale), cognitive processing (as measured by eye-tracking metrics), and learning outcomes (as measured by retention and transfer tests). The term emotional design refers to the redesign of the graphics in a multimedia material to improve the visual appeal and anthropomorphism (or personification) of the key elements in the lesson (Mayer & Estrella, 2014; Plass & Kaplan, 2016). Visual appeal involves rendering each element in a distinct, appealing color and anthropomorphism involves rendering essential elements with human-like features, such as facial expression (Mayer & Estrella, 2014; Plass & Kaplan, 2016). For example, consider a short instructional video that explains how lightning works and contains the essential elements of water droplets and ice crystals in black-and-white tone, as exemplified in the left top of Figure 1 (no emotional design).

Figure 1.

Screenshots of essential elements in four versions of the instructional videos.

In an effort to enhance the visual appeal and anthropomorphism of the essential elements, we redesigned the graphics of water droplets and ice crystals with appealing colors (e.g., blue), as exemplified in the right top of Figure 1 (colorful emotional design), face-like characters (e.g., smiling eyes and mouths), as exemplified in the left bottom of Figure 1 (anthropomorphizing emotional design), and both of them, as exemplified in the right bottom of Figure 1 (colorful and anthropomorphizing emotional design). In short, our goal is to determine whether the colorful and anthropomorphizing emotional design treatment is more effective than the colorful emotional design or anthropomorphizing emotional design treatments in promoting affective processes, cognitive processes, and learning outcomes.

The rationale of the present study is that the redesigned graphics are intended to induce learners’ positive emotions (i.e., affective processing during learning), and thereby increase their cognitive engagement in making sense of the essential material in the lesson (i.e., cognitive processing during learning), which leads to improved learning outcomes. In the present study, affective processing is assessed through self-report surveys of experienced positive emotion during learning, cognitive processing is assessed through eye-tracking metrics during learning, and learning outcome are assessed through retention and transfer posttests.

Role of Emotion in Multimedia Learning

This study reflects growing interest in the role of affect in multimedia learning (Schrader et al., 2022). In particular, this study examines the role of instructional design features intended to promote positive emotion in learners during a lesson, which can be called emotional design (Mayer & Estrella, 2014; Plass & Kaplan, 2016; Plass et al., 2014; Um et al., 2012). Research on emotional design fits within the broader issue of the role of emotion in academic learning (Pekrun, 2017; Pekrun & Linnenbrink-Garcia, 2012; Pekrun & Perry, 2014; Plass & Kalyuga, 2019; Tettegah & McCreery, 2015).

This study is grounded in Russell’s (1980, 2003) model of core affect, which posits that human emotion can be classified along two dimensions: valence (which runs from positive to negative) and arousal (which runs from active to passive). Research on emotion displayed by human and virtual instructors pinpoints the role of the valence dimension in academic learning, such as positive lessons involving happy or content emotion or negative lessons involving frustrating or boring emotion. In particular, the positivity principle holds that students can perceive the positive tone of the lesson, feel the positive tone of the lesson, work harder to learn with positive lessons, and learn better with positive lessons (Horovitz & Mayer, 2021; Lawson et al., 2021a; Lawson & Mayer, 2021a).

Emotional Design in Multimedia Learning

In the past few years, several studies had shown that multimedia learning materials could be redesigned in a way to induce learners’ positive emotions that in turn facilitate their learning (e.g., Chen & Wang, 2011; Lawson et al., 2021a; Lawson & Mayer, 2021a; Mayer & Estrella, 2014; Plass et al., 2014; Um et al., 2012). One approach to inducing learners’ positive emotions is emotional design. Most of the studies on emotional design in multimedia learning materials employed two main design techniques: incorporating appealing colors and anthropomorphic elements. The first known research on emotional design in multimedia was conducted by Um et al. (2012). They redesigned the main elements of a multimedia lesson on how immunization works from gray-tone rectangular shapes (neutral emotional design) to round shapes with face-like expressions rendered in bright warm colors (positive emotional design). The results showed that compared to the neutral emotional design, positive emotional design successfully induced learners’ positive emotions as measured by the Positive Affect Scale (PAS) and improved their learning outcomes as measured by comprehension and transfer tests.

Plass et al. (2014) replicated Um et al.’s (2012) study and observed that both face-like shapes alone and in combination warm colors helped induced students’ positive emotions. In addition, they found that the students who learned the multimedia lesson with emotional design reported less perceived difficulty and performed better on a comprehension test but not on a retention test. Mayer and Estrella (2014) followed up with a similar study on emotional design involving college students and found that the emotional design group performed better than the neutral design group on a retention test but not on a transfer test, but they did not directly investigate whether the emotional design induced positive emotions during learning. Le et al.’s (2018) study also found that compared to the neutral design group who learned a biology multimedia lesson with monochromatic grayscale and rectangular shapes, the positive emotional design group who learned the multimedia lesson with saturated and analogous warm color and face-like round shapes performed better on retention test and had a stronger decrease in the high-frequency band of heart rate variability which reflected a greater mental effort investment.

Following these foundational studies, the research literature on emotional design of multimedia lessons has begun to expand. Brom et al. (2018) carried out a meta-analysis of 33 independent experimental comparisons; the results indicated that adding anthropomorphic faces and pleasant colors in multimedia learning materials were effective emotional design approaches that tended to improve retention test scores on (g = 0.387 based on 18 comparisons), transfer (g = 0.327 based on 27 comparisons), and comprehension (g = 0.317 based on 14 comparisons). In a more recent follow-up meta-analysis, Wong and Adesope (2020) also found that adding emotional design elements to a multimedia lesson tended to improve retention (g = 0.35 based on 28 comparisons), transfer (g = 0.27 based on 38 comparisons), and comprehension test scores (g = 0.29 based on 16 comparisons).

Yet, some studies have also shown that emotional design did not induce positive emotions or improve learning outcomes. For instance, Park et al. (2015) used eye-tracking technology to examine the influence of emotional design on multimedia learning. The results showed that anthropomorphism had an effect of capturing attention, but had no effect on increasing positive emotions or learning outcomes. Uzun and Yıldırım (2018) explored the effects of four different amount of emotional design features: no emotional design features (Neutral Design, ND), adding bright and saturated colors (Colorful Design, CD), adding expressive facial expressions (Anthropomorphic Design, AD), and adding interesting sound effects design features (Anthropomorphic Design and Sound Effects group, ADSE). The results indicated that learners’ positive emotions generally increased as the amount of emotional design features in multimedia lesson increased, but no significant differences were observed in the transfer test performance among the four groups. These study findings indicated that adding emotional design features does not always lead to improved learning outcomes.

Overall, previous studies of the emotional design in multimedia learning materials have not examined whether adding two emotional design features (e.g., appealing colors and anthropomorphic elements) has a greater effect than adding one feature (e.g., appealing colors or anthropomorphic elements) on affective processes and cognitive processes, and questions remain as to how combinations of emotional design features impact affective processes and cognitive processes. Therefore, in the present study, we aim to expand the previous studies by taking a deeper and more systematic approach. Based on existing studies, we focused on two emotional design features in this study: appealing colors and anthropomorphic. We used as self-report questionnaire to gauge the learner’s felt emotions during learning, as an indication of affective processing during learning. In addition, we used eye-tracking technology to measure learners’ eye movements during learning, as an indication of cognitive processing during learning. We explored the role that these measures of affective and cognitive processes play as internal mechanisms by which emotional design causes learners to gain better learning outcomes.

Using Eye-tracking Technology to Understand Emotional Design in Multimedia Learning

An important feature of the present study is the use of measures of affective processes during learning (via self-report surveys), cognitive processes during learning (via eye-tracking technology), and learning outcomes (via open-ended retention and transfer posttests) all in one study. In particular, in order to examine cognitive processing during learning, we employ eye-tracking technology, which allows researchers to investigate learners’ allocations of visual attention (Holmqvist et al., 2011; Jarodzka, 2022; Lai et al., 2013). In particular, the eye movements (e.g., fixation duration and fixation count) captured through eye-tracking devices show what elements of a stimulus learners are looking, for how long, and in what order (Zhai et al., 2018). Emotional design in multimedia learning materials have been shown to guide visual attention (Park et al.,2015; Peng et al., 2021; Stárková et al., 2019). As such, eye-tracking technology is beginning to be used to determine the impact of emotional design on learning, which is not easily obtained through the traditional learning assessments such as self-reports or tests.

As mentioned above, Park et al. (2015) conducted an eye-tracking study on emotional design in multimedia learning. Their study results revealed that anthropomorphism in the multimedia learning material captured learners’ attention as indicated by the longest fixation duration on the relevant information embedded in the emotional design elements. Stark et al. (2018) used eye-tracking technology to investigate the influence of text-based emotional design in a multimedia instruction. They found that neither the positive emotional text design nor the negative emotional text design primed more attention on the text passages than the neutral text as measured by fixation time. The reason may be that the variations in the texts caused by emotional design are too small to alter learners’ eye movements. Stárková et al. (2019) found that participants were quicker to locate the emotionally designed anthropomorphic graphics as indicated by initial dwell time. In this study, we explored the value of eye-tracking technology as a tool to examine the impacts of emotional design of visual content on students’ cognitive processing during learning.

Theory and Predictions

This study is grounded in the Cognitive-Affective Model of E-Learning (Lawson et al., 2021a; 2021b; Lawson & Mayer, 2021a; 2021b; Mayer, 2020) and the Cognitive Affective Theory of Learning with Media (CATLM; Moreno, 2007; Moreno & Mayer, 2007) from which it is derived. The Cognitive-Affective Model of E-Learning (Mayer, 2020) posits a causal chain in which a lesson displays positive emotion, the learner perceives and feels positive emotional tone (i.e., affective processing), the learner exerts effort to learn from the lesson, (i.e., cognitive processing) and the learner performs well on learning outcome tests. Figure 2 shows a simplified rendering of the model, which is examined in the present study.

Figure 2.

A cognitive-affective model of e-learning.

Concerning the first box in Figure 2, in line with the positivity principle, we sought to create positive tone in a multimedia lesson by giving key elements in the lesson positive facial expressions and by using appealing colors. We chose to enhance our lesson with positive facial expression because this technique has been hypothesized to convey positive emotional tone when displayed by instructors giving a lecture in an instructional video (Horovitz & Mayer, 2021; Lawson et al., 2021a; 2021b; Lawson & Mayer, 2021a) and in online multimedia lessons (Mayer & Estrella, 2014; Plass et al., 2014; Um et al., 2012). We chose to enhance our lesson with appealing colors because they have been hypothesized to convey positive emotional tone in online multimedia lessons and educational games (Plass & Kaplan, 2016; Plass et al., 2014; 2020; Um et al., 2012).

Concerning the second box in Figure 2, in the present study, we sought to determine whether the added features affected learners' affective processing by causing them to experience positive emotion during the lesson. We measured learners' affective processing through a self-report questionnaire focusing on the learner’s experienced positive emotion (i.e., mean rating on the Positive Affect Scale). We interpret this metric as an index of the learner’s level of experienced positive emotion during learning.

Concerning the third box in Figure 2, in the present study, we sought to determine whether the added features affected learners' cognitive processing by causing them to focus their attention on the relevant elements on the screen. We measured learners' cognitive processing through eye-tracking metrics (i.e., time to first fixation on emotional design AOIs and fixation duration on emotional design AOIs). We interpret these metrics as an index of the level of the learner’s attending to the relevant portions of the graphic in the lesson, which indicates that the learner’s cognitive processing is on task.

Finally, concerning the fourth box in Figure 2, in the present study, we sought to determine whether the added features affected learning outcomes, as measured by retention and transfer tests. If adding positive elements to the lesson mainly serves to guide the learner’s attention, then we expect the treatment groups to outperform the control groups on the retention test. If adding positive elements to the lesson mainly helps learners to process the material more deeply by mentally reorganizing the material and relating it to relevant prior knowledge, then we expect the treatment groups to outperform the control group on the transfer test. It is not clear whether the effects of the two techniques for conveying positive emotion are equivalent (enabling one technique for conveying positive emotion in a lesson to substitute for the other) or whether the effects are additive. We address this exploratory issue in the present study by comparing groups that receive anthropomorphism emotional design, colorful emotional design, both, or neither.

Based on the positivity principle as represented in the Cognitive-Affective Model of E-Learning, we focus on several hypotheses:

Hypothesis 1

The anthropomorphism emotional design, colorful emotional design, and colorful & anthropomorphizing emotional design groups will provide higher mean ratings of experienced positive emotion than the control group.

Hypothesis 2

The anthropomorphism emotional design, colorful emotional design, and colorful & anthropomorphizing emotional design groups will exhibit faster time to first fixation on emotional design AOIs and greater fixation duration on emotional design AOIs than the control group.

Hypothesis 3

The anthropomorphism emotional design, colorful emotional design, and colorful & anthropomorphizing emotional design groups will outperform the control group on the retention and transfer tests.

Hypothesis 4

If the effects of the two positivity treatments are equivalent, then we predict that the anthropomorphism emotional design, colorful emotional design, and colorful & anthropomorphizing emotional design groups will be equivalent in outperforming the control group on positive emotion ratings, eye-tracking scores, and the retention and transfer tests (hypothesis 4a). If the effects of the two positivity treatments are additive, then we predict that the colorful & anthropomorphizing emotional design group will outperform the control group on positive emotion rating, eye-tracking scores, and the retention and transfer tests, whereas the anthropomorphism emotional design and colorful emotional design groups will not (hypothesis 4b).

Hypothesis 5

Concerning the mechanism of change, we wish to investigate possible pathways between the level of emotional design in a lesson and scores on the retention and transfer test in order to better understand the role of affective and cognitive processes during learning as possible mediators. Figure 3 presents a model of the hypothesized relationships based on the Cognitive-Affective Model of E-Learning (Mayer, 2020), and on the studies that had identified the impacts of emotional design on multimedia learning (Moreno, 2007; Moreno & Mayer, 2007). The model outlines how emotional design influences positive emotion (link 1) and fixation duration (link 2). Positive emotion further is related to the cognitive process indicated by fixation duration (link3), and to the learning outcomes indicated by retention (link 4) and transfer (link 5). Fixation duration is further related to retention (link 6) and transfer (link 7). Finally, we also model the relationship between retention and transfer (link 8, 9). According to the model of hypothesized relationships in Figure 3, we predicted that affective processes (i.e., positive emotion) and cognitive processing (i.e., fixation duration) will mediate the effects of emotional design on learners’ learning outcomes (i.e., retention test and transfer test scores).

Figure 3.

Hypothesized relationships based on the cognitive-affective model of e-learning.

Method

Participants and Design

The participants in the present experiment were 100 college students recruited from a teachers’ university in Tianjin, China. The participants were from 18 to 25 years old (M = 19.31, SD = 1.81). There were 58 women and 42 men. All the participants were native Chinese speakers and had normal or corrected to normal vision.

In a between-subjects design, the participants were randomly assigned to one of the four experimental groups: no emotional design group, colorful emotional design group, anthropomorphism emotional design group, and colorful & anthropomorphism emotional design group. Twenty-five participants served in each group. Consistent with the procedure in several previous studies (e.g., Park et al.,2015; Wang et al., 2020), firstly the male participants were randomly assigned to the four groups, and then the female participants were randomly assigned to the four groups, to ensure that the proportion of males and females in each group was the same (i.e., about 10–11 males and 14–15 female students out of a total of 25 students in each condition). We used this kind of random assignment to avoid the interference of gender factors on the experimental results. A chi-square test result showed that there was no difference among the groups in proportion of male and female students, p = .983. Analyses of variance (ANOVAs) indicated that there were no significant differences among the groups on prior knowledge, F (3, 96) = 1.219, p = .307, and age, F (3, 96) = .423, p = .737.

Learning Materials

The multimedia materials consisted of four versions of a short instructional video regarding how lightning develops and works (Mayer, 2021; Mayer & Moreno, 1998; Moreno & Mayer, 2002). The short instructional video lasted for 2 minutes and 45 seconds with Chinese narration. The lesson explained the process of lightning formation, including the role of differences in air temperature and differences in electrical charge. The no emotional design version portrayed the graphic elements in the lesson in black and white line drawings, with water droplets portrayed as a tear-shapes, ice crystals portrayed as an asterisk, each positively-charged particle as a plus sign within a circle, and each negatively-charged participle a minus sign with a circle. Thus, in the no emotional design version all graphics were in black-white tone and there were no emotional design features added in the essential graphic elements. The three versions of emotional design instructional video were identical to the control version of no emotional design version except the essential graphic elements (e.g., water droplets, ice crystals, positive charge, and negative charge) were redrawn. In the colorful emotional design version, we used vibrant and bright colors, such as blue for water droplets and ice crystals, red for positive charge, purple for negative charge, and so on, in an attempt to attract the students’ attention to the essential elements in the video with positive emotions. In the anthropomorphism emotional design version, we added face-like characters, such as smiling eyes and mouths, into the essential elements in an attempt to induce the students’ positive emotions. In the colorful & anthropomorphism emotional design version, both vibrant and bright colors and face-like characters were added in the essential elements to enhance the visual appeal and to induce students’ positive emotions. Figure 1 shows screenshots of essential elements from the four versions of the instructional video.

Devices

A Tobii X120 eye tracker with a sampling rate of 120 Hz was used to track eye movements while each of the participants learned the instructional video. As the Tobii X120 is a remote eye tracker, each participant could sit about 50 cm in front of the eye tracker while having their head positioned at a natural distance from the screen and moving their head freely without paying attention to the eye tracker. This setting minimized device interference with learner’s cognitive processes. Tobii Studio 3.1.6 software was used to perform the calibration process and analyze the eye movement data. We also recorded brain activity using a CUBand EEG device, which consisted of a headband with a dry electrode, but we did not use the brainwave data in the present analyses due to technical problems.

Measures

The measures tapped prerequisite learner characteristics, positive emotion rating scores, eye-tracking metrics, and learning outcome scores.

Prerequisite learner characteristics were solicited from the demographic questionnaire and prior knowledge questionnaire. The demographic questionnaire collected information concerning the participant’s gender and age (Mayer & Moreno, 1998). The prior knowledge questionnaire consisted of four five Likert-scale items in which students rated their knowledge about the formation of cloud; the movement of electric charge; gasification, liquefaction and solidification; and how lightning works (Cronbach’s α = .814). Participants were asked to assess their knowledge about meteorology ranging from 1 (know nothing) to 5 (know well). The prior knowledge score consisted of the mean rating across the four items. Participants who reported high prior knowledge experiences (i.e., greater than an average of 3 point) were excluded from the experiment. We used the self-rate familiarity with the video lesson as the measurement of prior knowledge rather than a pretest of the video lesson to avoid creating a priming effect or testing effect in which taking a pretest is a learning part that also guides the learner’s attention during the video lesson (Mayer, 2021).

The participants’ experienced positive emotion was assessed using the (PAS, Cronbach’s α = .899) from Watson et al.’s (1988) Positive and Negative Affect Schedule. The PAS is an established measure of positive emotion that has been successfully applied in previous researches (e.g., Crawford & Henry, 2004; Plass et al., 2020; Riva et al., 2007; Um et al., 2012). The participants reported the extent to which they experienced 10 different feelings related to positive emotion (e.g., excited, enthusiastic, interested), using a five Likert-scale ranging from 1(very slightly or not at all) to 5 (very much). The possible total score of positive emotion was from 10 to 50.

Concerning eye-tracking metrics, we defined two kinds of areas of interest (AOIs): the emotional design AOIs and the non-emotional design AOIs. Figure 1 shows the emotional design AOIs, which consisted of essential elements in the graphics that had emotional design features applied to them such as color and/or personification. The non-emotional design AOIs, which consisted of non-essential elements that did not have emotional design features applied to them. We derived three kinds of eye-tracking metrics: time to first fixation on an emotional design AOI, fixation duration on emotional design AOIs, and fixation duration on non-emotional design AOIs (Holmqvist et al., 2011). Time to first fixation on an emotional design AOI was the amount of time that each participant took to fixate on an emotional design AOI for the first time, which indicates the speed of visual search for emotional design elements in the lesson (Gillespie-Smith et al., 2016). We considered this time to first fixation on an emotional design AOI as representing the impact that the emotional design had on catching the learners’ attention. Fixation duration on emotional design AOIs is the total amount of time of all the fixation durations in emotional design AOIs, whereas fixation duration on non-emotional design AOIs is the total time of all durations in non-emotional design AOIs. Researchers most commonly used fixation duration as measure for information selection during learning processes in previous studies (e.g., Ponce & Mayer, 2014a; 2014b; Wang et al., 2020). Similarly, Park et al.’s (2015) study showed that the anthropomorphism elements promoted attentional engagement resulting in longer fixation duration on the emotional design graphics AOIs. We consider fixation duration on emotional design AOIs and fixation duration on non-emotional design AOIs as representing the impact that emotional design had on the allocation of learners’ cognitive processing resources.

Concerning the learning outcome, two learning outcome measures were used in this study: retention test score and transfer test score. The retention test and transfer test were developed based on Harp and Mayer’s (1997) study. The retention test (Cronbach’s α = .827) used seven single-choice questions to evaluate to what extent the students remembered factual knowledge that was directly stated in the instructional video. One point was given for each correct choice, yielding possible scores from 0 to 7. The seven questions were about the formation of clouds, the formation of ice crystals, the formation of electrical charges in clouds, the movement of charges in clouds, the cascade pilot lightning stroke, the return lightning stroke, and the lightning channel. The transfer test was used to evaluate to what extent the students could apply the presented knowledge to solve new problems that were not explicitly stated in the instructional video. The transfer test (Cronbach’s α = .807) consisted of two open-ended questions. Five points was given for each good answer, yielding possible scores from 0 to 10. The two open-ended questions were: 1) If you see clouds in the sky but no lightning, why is that? 2) How to reduce the harm of lightning? The learning outcome tests were separately scored by two trained physics graduates who were blind to the four experimental conditions. The scorers’ consistency coefficient was above 0.90. We used the average score of the two scorers’ scores as the learning outcome test score.

Procedure

The participants were randomly assigned to one of the four groups and tested individually in an eye movement lab. There were four steps in the present experiment (as summarized in Figure 4): pre-learning activities (consisting of the demographic and prior knowledge questionnaires), learning (consisting of eye tracking monitoring during learning the instructional video), post-learning query (consisting of the positive affective state rating), and learning outcome posttests (consisting of the retention and transfer tests).

Figure 4.

The procedure of the experiment.

First, concerning the pre-learning activities, participants were asked to filled out the demographic and prior knowledge questionnaires. Participants who rated their knowledge higher than 3 out of 5 were excluded from the experiment and replaced to maintain 25 participants per group. Second, concerning the learning phase, participants viewed the instructional video corresponding to their group, subject to eye tracking. Before the lesson, the experimenter explained the eye tracking technology to the participants. The eye-tracking calibration process was performed. We excluded participants whose eyes could not be calibrated (n = 5). Third, concerning the post-learning query, participants completed the Positive Affect Scale (PAS) at their own rate. Fourth, concerning the learning outcome posttests, participants were asked to complete the retention and transfer tests with a 20-minute time limit.

We obtained IRB approval and adhered to ethical guidelines for conducting research with human subjects.

Results

Statistical Analyses

Since the type of emotional design (colorful, anthropomorphism, colorful & anthropomorphism, or none) was the only between-subjects factor manipulated in this experiment, we conducted one-way ANOVAs to investigate the differences among the groups in positive emotion scores, eye movement metrics, and learning outcome posttest scores. Dunnett’s tests (at p < .05) were conducted as post-hoc planned comparisons to analyze the predictions of this study, that is, comparing each of emotional design groups against the no emotional design group. Structural equation modeling (SEM) analyses in Amos 24 were conducted to investigate how learners’ positive emotions and eye movements are mediate the effects of emotional design on learning outcomes.

Hypothesis 1: Does Adding Emotional Design Cause Increases in Affective Processing during Learning as Indicated by Higher Ratings of Learners’ Positive Emotion?

The first hypothesis is that adding positive emotional design elements to a multimedia lesson will evoke positive emotion in the learner. The first row of Table 1 shows the means and standard deviations of the learners’ positive emotion ratings on the PAS. A one-way ANOVA showed that the four groups significantly differed on positive emotion rating scores, F (3, 96) = 3.163, p = .028, η² = .090. A Dunnett’s test showed that the anthropomorphizing emotional design group (d = .804) and colorful & anthropomorphizing emotional design group (d = .726) scored significantly higher on positive emotion than the no emotional design group, whereas the colorful emotional design group did not (d = .399). These results partially support hypothesis 1, suggesting that positive emotional design—either in the anthropomorphizing emotional design or colorful & anthropomorphizing emotional design group—was helpful for evoking learners’ positive emotions.

Table 1.

Means and Standard Deviations on the Measures for Each Group.

Measures	No emotional design group		Colorful emotional design group		Anthropomorphizing emotional design group		Colorful & anthropomorphizing emotional design group
Measures	M	SD	M	SD	M	SD	M	SD
Positive emotion	25.92	6.58	28.68	7.21	31.16	6.45	30.56	6.19
Time to first fixation on an emotional design AOI (s)	1.92	2.11	1.42	1.79	1.39	2.03	0.56	0.59
Fixation duration on emotional design AOI (s)	25.41	8.92	28.15	5.28	29.59	7.49	30.46	6.69
Fixation duration on non-emotional design AOIs (s)	85.20	24.07	91.40	13.20	92.30	16.27	89.35	17.01
Retention	4.56	1.15	4.60	1.19	4.48	1.29	4.76	1.01
Transfer	2.44	4.65	5.56	6.13	3.48	4.74	6.04	5.87

Note. Scores in bold font are significantly greater than for the no emotional design group (at p < .05).

Hypothesis 2: Does Adding Emotional Design Cause Increases in Cognitive Processing During Learning as Measured by Eye-tracking Metrics?

The second hypothesis is that adding positive emotional design elements to a multimedia lesson will cause changes in eye-tracking metrics (i.e., a decrease in time to first fixation on an emotional design AOI and an increase in fixation duration on emotional design AOIs). The second row of Table 1 shows the means and standard deviations on the time to first fixation on an emotional design AOI. A one-way ANOVA showed that the differences on the time to first fixation on an emotional design AOI did not reach statistical significance, F (3, 96) = 2.586, p = .058, η² = .075. A Dunnett’s test showed that the colorful & anthropomorphizing emotional design group had a significantly shorter time to first fixation on an emotional design AOI than the no emotional design group (d = - .877), whereas the colorful emotional design (d = - .255) and anthropomorphizing emotional design (d = - .255) groups did not. These results partially support hypothesis 2, suggesting that the positive emotional design—in the colorful & anthropomorphizing emotional design group—was helpful for guiding learners’ attention to the emotional design areas.

The third row of Table 1 shows the means and standard deviations on fixation duration on emotional design AOIs. A one-way ANOVA showed that the differences in fixation duration on emotional design AOIs did not reach statistical significance, F (3, 96) = 2.344, p = .078, η² = .068. A Dunnett’s test showed that the colorful & anthropomorphizing emotional design group had a significantly longer fixation duration on emotional design AOIs than the no emotional design group (d = .640), whereas the colorful emotional design (d = .373) and anthropomorphizing emotional design (d = .507) groups did not. These results partially support hypothesis 2, suggesting that positive emotional design—in the colorful & anthropomorphizing emotional design group—was helpful for focusing learners’ attention on the emotional design AOIs.

Although we had no predictions concerning differences in fixation duration on non-emotional design AOIs, we added this analysis for exploratory purposes. The fourth row of Table 1 shows the means and standard deviations on fixation duration on non-emotional design AOIs. As could be expected, a one-way ANOVA showed that there was no significant difference on fixation duration on non-emotional design AOIs, F (3, 96) = .764, p = .517, η² = .023.

Overall, the eye-tracking results encourage the idea that the combination of colorful and anthropomorphic emotional design elements can prime productive cognitive processes in learners aimed at attending to relevant visual information.

Hypothesis 3: Does Adding Emotional Cause Improvements in Learning Outcome as Measured by Retention and Transfer Tests?

The third hypothesis is that learners will perform better on retention and transfer tests if positive emotional design elements are added to their multimedia lesson. The bottom two rows of Table 1 show the means and standard deviations for each group on retention and transfer test scores, respectively. A one-way ANOVA revealed that there was no significant difference on the retention test scores, F (3, 96) = .254, p = .858, η² = .008, and the difference on the transfer test scores also did not reach statistical significance, F (3, 96) = 2.499, p = .064, η² = .072, among the four groups. A Dunnett’s test showed that the colorful emotional design (d = .573) and colorful & anthropomorphizing emotional design (d = .679) groups significantly outperformed the no emotional design group on transfer scores, whereas the anthropomorphizing emotional design group did not (d = .221). These results partially support hypothesis 3, suggesting that positive emotional design—either in the colorful emotional design or colorful & anthropomorphizing emotional design group—was helpful for promoting knowledge transfer. The reasons for the emotional design did not promote knowledge memory may be that deeper processing is expected to have strongest effects on transfer rather than retention (Mayer, 2011).

Hypothesis 4: Are the Effects of Adding Anthropomorphism and Colorization Additive?

Across all three types of measures—affective processing (via emotion rating), cognitive processing (via eye-tracking metrics), and learning outcome (via transfer test score)—the group that received both emotional design features (colorful & anthropomorphizing emotional design) outperformed the control group, but the groups that received one emotional design feature (anthropomorphizing emotional design or colorful emotional design) did not. This pattern is most consistent with hypothesis 4b, indicating that the combination of two emotional design features is stronger than one feature alone.

Hypothesis 5: Does Affective and Cognitive Processing Mediate the Effects of Positive Emotional Design on Learning Outcomes?

Concerning the mechanisms of change by which emotional design can influence learning outcomes, hypothesis 5 is that affective processes (indicated by emotion rating) and cognitive processes (indicated by eye-tracking metrics) mediate the effect of adding emotional design features (i.e., emotional design vs. no emotional design) on learning outcome scores (indicated by retention and transfer test scores), as proposed in Figure 3. In order to investigate this model, we used SEM to calculate the fit of the data to the hypothesized relationships proposed in Figure 3 (Kim & Nembhard, 2019; Lin et al., 2019). Since compared to the no emotional design group, the colorful & anthropomorphism emotional design group showed the higher positive emotion rating, the shortest time to first fixation on an emotional design area, the longest fixation duration on emotional design areas, and the best transfer test score, we used the data of the colorful & anthropomorphizing emotional design and no emotional design groups for calculating the model fit. The results showed that the Chi-square (χ²) = 2.675 with df =3, the p-value was 0.445, which was greater than the recommended 0.05 (Chen & Lin, 2010), the CFI was 0.99, which was greater than the recommended 0.96 (Hooper et al., 2008; Kim & Nembhard, 2019), and the RMSEA was 0.01, which was lower than the recommended 0.05 (Lai & Green, 2016). Figure 5 illustrates the relationships with the significant non-standardized estimate path coefficients.

Figure 5.

The relationship model that only includes the significant paths in thick lines.

First, the path from emotional design condition to positive emotion was significant (β = 4.640, p = .010, se = 1.791), which indicated that the students in the emotional design condition—colorful & anthropomorphizing emotional design—reported higher positive emotions than those in the no emotional design condition as expected. Furthermore, positive emotion was a significant antecedent of retention test score (β = .590, p = .001, se = .180) and transfer test score (β = .448, p < .001, se = .131), as predicted. Thus, this path showed how the effects of emotional design on retention and transfer scores were mediated by learners’ positive emotions (i.e., a measure of affective processing during learning).

Second, the path from emotional design condition to fixation duration (i.e., fixation duration on emotional design AOIs) was significant (β = 5.049, p = .022, se = 2.209), and fixation duration was a significant antecedent of retention test score (β = .305, p < .001, se = .077), but not a significant antecedent of transfer test score. Thus, this path showed how the effects of emotional design on retention test score were mediated by fixation duration (i.e., a measure of cognitive processing during learning).

Overall, this pattern partially supports hypothesis 5 by revealing two pathways from emotional design in multimedia lesson to learning outcomes—the positive emotion pathway and the fixation duration pathway. In the positive emotion pathway, emotional design in multimedia lesson, which is related to learners’ positive emotion, which in turn is related to retention test and transfer test scores. In the fixation duration pathway, emotional design in multimedia lesson, which is related to learners’ fixation duration, which in turn is related to retention test score. This work gives a preliminary picture of possible pathways from emotional design to learning outcomes involving affective processes and cognitive processes as mediating factors.

Discussion

Empirical Contributions

This study examined the effects of emotional design of key visual elements in a multimedia lesson—colorful emotional design, anthropomorphism emotional design, and colorful & anthropomorphism emotional design—on measures of affective processing during learning (measured by positive emotion ratings), cognitive processing during learning (measured by eye-tracking metrics), and learning outcomes (measured by retention and transfer test scores). Compared to no emotional design, colorful & anthropomorphizing emotional design resulted in higher ratings of positive emotion (d = .726) as a measure of affective processing, longest fixation duration on emotional design AOIs (d = .640) and shortest time to first fixation on an emotional design AOI (d = - .877) as measures of cognitive processing, and best transfer performance (d = .679) as a measure of learning outcome. Across all three types of measures—affective processing, cognitive processing, and learning outcome, the colorful & anthropomorphizing emotional design outperformed the control group, but the groups that received one emotional design feature (anthropomorphizing emotional design or colorful emotional design) did not. Finally, a structural equation model showed that the effects of emotional design in learning performance are mediated by positive emotion and fixation duration, thereby highlighting the role of affective and cognitive processes in multimedia learning. This work helps extend findings by Kumar et al. (2019), Le et al. (2018), Park et al. (2015), Plass et al. (2014, 2020), Uzun and Yıldırım (2018), and Um et al. (2012) by using a different lesson, testing college students in a different nation, and including measures of the mechanisms of change by which emotional design can influence learning outcomes. The major empirical contribution of this work is to show that adding two emotional design elements to a multimedia lesson creates positive effects on measures of affective processing, cognitive processing, and learning outcomes. Overall, adding two emotional design features was consistently effective whereas adding one was not.

Theoretical Implications

Overall, the findings of the present study are consistent with the predictions of the Cognitive-Affective Model of E-Learning (Lawson et al., 2021a; 2021b; Lawson & Mayer, 2021a; 2021b; Mayer, 2020) and the Cognitive Affective Theory of Learning with Media (Moreno, 2007; Moreno & Mayer, 2007), which posits that emotional design—making essential elements visually appealing and salient (e.g., through color and anthropomorphism)—evokes affective processing by inducing positive emotion in learners and guides cognitive processing during learning by directing attention to better understand the key elements.

Concerning learning outcome measures, the hallmark of the effects of emotional design on learning is improvements in transfer test performance. Concerning positive emotion measures, the hallmark of the effects of emotional design on learning is an increase in ratings of positive emotion. Concerning eye-tracking measures, the hallmark of the effects of emotional design on learning is an increase in the fixation duration on emotional design elements and a decrease in searching time to emotional design elements. Concerning the mechanisms of change by which emotional design can influence learning outcomes, the hallmark of the effects of emotional design on learning is a mediation by positive emotion and fixation duration in learning performance. The major theoretical contribution of this work is that the effects of emotional design on learning outcomes are mediated by affective and cognitive processes during learning.

Practical Implications

This study points to the potential of combining two emotional design elements—colorizing and anthropomorphizing—in a multimedia lesson as an instructional design method for evoking positive emotion, and guiding and maintaining cognitive processing during multimedia learning, which results in better learning outcomes. This study shows that the most powerful and consistent way to create positive emotional design to combine both appealing color and anthropomorphism for essential graphical elements in a multimedia lesson. Based on the results of this study, we recommend that instructional designers should identify the key graphical elements in a lesson rather than add irrelevant graphical elements and render them in appealing colors with positive human-like facial expressions. In addition, this study shows that although anthropomorphism for key graphical elements did not produce better learning outcomes, it induced learners’ positive emotions, and although appealing color for key graphical elements did not induce learners’ positive emotions, it produced better learning outcomes. The potential benefits of combining colorful emotional design and anthropomorphism emotional design should not be ignored by instructional designers. On that basis, we recommend that instructional designers use both emotional design elements in multimedia lessons, while considering distinct learner characteristics and needs.

Conclusion, Limitations, and Future Directions

Overall, the findings of the present study suggest that adding two emotional design techniques (colorful & anthropomorphism) to the key graphical elements in a multimedia lesson creates more positive effects on affective processes, cognitive processes, and learning outcomes than adding one (colorful or anthropomorphism).

There are several limitations of this study. First, the learning material was a short video, lasting only 2 minutes and 45 seconds. We used a short lesson in the consideration of the experimental cycle and the participants’ potential fatigue. Further research should find employ longer lessons. Second, the learning materials used in this study were based on one subject area—the physics of lightning storms. A more diverse range of learning materials can be used for future research to examine if similar research results would emerge. Third, this study used a delayed questionnaire as measurement to investigate learners’ positive emotion during learning. Future research should use direct measures of learners’ emotion during learning. Fourth, although this study examined the impacts of appealing color versus black and white on emotions, future research is needed to explore the influence of different color tones on emotions. For example, cold colors like blue may have an impact on emotion that is different from warm colors. Fifth, the participants in this study are 100 college students, further research should recruit larger numbers of participants in different grades to examine the findings. Finally, the EEG measures were not useful in the present study, so future research should use more sensitive brain-based measures, possibly involving more electrodes.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (NSFC), Grant number: 62107030.

Statements on Open Data,and Ethics

In this research, all participants took part in the experiment voluntarily. Data were collected and used after participants granted their consent. Personal or personally identifiable information was not included in the data.

ORCID iDs

Xue Wang

Richard E. Mayer

Meiqi Han

Lei Zhang

Author Biographies

Xue Wang is an associate professor in the Faculty of Education, Tianjin Normal University, Tianjin, China. Her research interests include multimedia learning, learning resources in online learning environments, linguistics for multimedia design, and learning analytics of online learning.

Richard E. Mayer is Distinguished Professor of Psychological and Brain Sciences at the University of California, Santa Barbara (USA). His research interests involve applying the science of learning to education, with a focus on multimedia learning, online learning, game-based learning, and learning in virtual reality.

Meiqi Han is a faculty member in the Office of Educational Administration, Tianjin Electronic Information College, Tianjin, China. Her research interests include multimedia learning, online learning, and learning analysis.

Lei Zhang is a postgraduate student in the Faculty of Education, Tianjin Normal University, Tianjin, China. Her research interests include multimedia learning, online learning, and learning analysis.

References

Brom

Stárková

D’Mello

S. K.

(2018). How effective is emotional design? A meta-analysis on facial anthropomorphisms and pleasant colors during multimedia learning. Educational Research Review, 25, 100−119. https://doi.org/10.1016/j.edurev.2018.09.004.

Chen

Lin

L.-S.

(2010). Structural equation-based latent growth curve modeling of watershed attribute-regulated stream sensitivity to reduced acidic deposition. Ecological Modelling, 221(17), 2086−2094. https://doi.org/10.1016/j.ecolmodel.2010.05.010.

Chen

C.-M.

Wang

H.-P.

(2011). Using emotion recognition technology to assess the effects of different multimedia materials on learning emotion and performance. Library & Information Science Research, 33(3), 244−255. https://doi.org/10.1016/j.lisr.2010.09.010.

Crawford

J. R.

Henry

J. D.

(2004). The positive and negative affect Schedule (PANAS): Construct validity, measurement properties and normative data in a large non‐clinical sample. British Journal of Clinical Psychology, 43(3), 245−265. https://doi.org/10.1348/0144665031752934.

Gillespie-Smith

Boardman

J. P.

Murray

I. C.

Norman

J. E.

O’Hare

Fletcher-Watson

(2016). Multiple measures of fixation on social content in infancy: evidence for a single social cognitive construct? Infancy, 21(2), 241−257. https://doi.org/10.1111/infa.12103.

Harp

S. F.

Mayer

R. E.

(1997). The role of interest in learning from scientific text and illustrations: On the distinction between emotional interest and cognitive interest. Journal of educational psychology, 89(1), 92−102. https://doi.org/10.1037/0022-0663.89.1.92.

Holmqvist

Nyström

Andersson

Dewhurst

Jarodzka

Van de Weijer

(2011). Eye tracking: A comprehensive guide to methods and measures. Oxford University Press.

Hooper

Coughlan

Mullen

M. R.

(2008). Structural equation modelling: guidelines for determining model fit. Electronic Journal of Business Research Methods, 6(1), 53−60.

Horovitz

Mayer

R. E.

(2021). Learning with human and virtual instructors who display happy or bored emotions in video lectures. Computers in Human Behavior, 119, 106724. https://doi.org/10.1016/j.chb.2021.106724.

10.

Jarodzka

(2022). Research methods in multimedia learning. In Mayer

R. E.

Fiorella

(Eds.) The Cambridge handbook of multimedia learning (3rd ed., pp. 41−55). Cambridge University Press.

11.

Kim

J.-E.

Nembhard

D. A.

(2019). Eye movement as a mediator of the relationships among time pressure, feedback, and learning performance. International Journal of Industrial Ergonomics, 70, 116−123. https://doi.org/10.1016/j.ergon.2018.12.006.

12.

Kumar

J. A.

Muniandy

Wan Yahaya

W. A. J.

(2019). Exploring the effects of emotional design and emotional intelligence in multimedia-based learning: an engineering educational perspective. New Review of Hypermedia and Multimedia, 25(1-2), 57−86. https://doi.org/10.1080/13614568.2019.1596169.

13.

Lai

Green

S. B.

(2016). The problem with having two watches: Assessment of fit when RMSEA and CFI disagree. Multivariate Behavioral Research, 51(2-3), 220−239. https://doi.org/10.1080/00273171.2015.1134306.

14.

Lai

M.-L.

Tsai

M.-J.

Yang

F.-Y.

Hsu

C.-Y.

Liu

T.-C.

Lee

S. W.-Y.

Lee

M.-H.

Chiou

G.-L.

Liang

J.-C.

Tsai

C.-C.

(2013). A review of using eye-tracking technology in exploring learning from 2000 to 2012. Educational Research Review, 10, 90−115. https://doi.org/10.1016/j.edurev.2013.10.001.

15.

Lawson

A. P.

Mayer

R. E.

(2021b). The power of voice to convey emotion in multimedia instructional messages. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-021-00282-y.

16.

Lawson

A. P.

Mayer

R. E.

Adamo-Villani

Benes

Lei

Cheng

(2021a). The positivity principle: Do positive instructors improve learning from instruction video lectures? Educational Technology Research and Development, 69(6), 3101−3129. https://doi.org/10.1007/s11423-021-10057-w.

17.

Lawson

A. P.

Mayer

R. E.

Adamo-Villani

Benes

Lei

Cheng

(2021a). Do learners recognize and relate to the emotions displayed by virtual instructors? International Journal of Artificial Intelligence in Education, 31(1), 134−153. https://doi.org/10.1007/s40593-021-00238-2.

18.

Lawson

A. P.

Mayer

R. E.

Adamo-Villani

Benes

Lei

Cheng

(2021b). Recognizing the emotional state of human and virtual instructors. Computers in Human Behavior, 114, 106554. https://doi.org/10.1016/j.chb.2020.106554.

19.

Liu

Deng

Dai

D. Y.

(2018). Heart rate variability reflects the effects of emotional design principle on mental effort in multimedia learning. Computers in Human Behavior, 89, 40−47. https://doi.org/10.1016/j.chb.2018.07.037.

20.

Lin

C. J.

Prasetyo

Y. T.

Widyaningrum

(2019). Eye movement measures for predicting eye gaze accuracy and symptoms in 2D and 3D displays. Displays, 60, 1−8. https://doi.org/10.1016/j.displa.2019.08.002.

21.

Mayer

R. E.

(2011). Applying the science of learning. Pearson.

22.

Mayer

R. E.

(2020). Searching for the role of emotions in e-learning. Learning and Instruction, 70, 101213. https://doi.org/10.1016/j.learninstruc.2019.05.010.

23.

Mayer

R. E.

(2021). Multimedia learning (3rd ed.). Cambridge University Press.

24.

Mayer

R. E.

Estrella

(2014). Benefits of emotional design in multimedia instruction. Learning and Instruction, 33, 12−18. https://doi.org/10.1016/j.learninstruc.2014.02.004.

25.

Mayer

R. E.

Moreno

(1998). A split-attention effect in multimedia learning. Journal of Educational Psychology, 90(2), 312−320. https://doi.org/10.1037/0022-0663.90.2.312.

26.

Moreno

(2007). Optimising learning from animations by minimising cognitive load: Cognitive and affective consequences of signalling and segmentation methods. Applied Cognitive Psychology: The Official Journal of the Society for Applied Research in Memory and Cognition, 21(6), 765−781. https://doi.org/10.1002/acp.1348.

27.

Moreno

Mayer

R. E.

(2002). Verbal redundancy in multimedia learning: When reading helps listening. Journal of Educational Psychology, 94(1), 156−163. https://doi.org/10.1037/0022-0663.94.1.156.

28.

Moreno

Mayer

(2007). Interactive multimodal learning environments. Educational Psychology Review, 19(3), 309−326. https://doi.org/10.1007/s10648-007-9047-2.

29.

Park

Knörzer

Plass

J. L.

Brünken

(2015). Emotional design and positive emotions in multimedia learning: An eye tracking study on the use of anthropomorphisms. Computers & Education, 86, 30−42. https://doi.org/10.1016/j.compedu.2015.02.016.

30.

Pekrun

(2017). Achievement emotions. In Elliot

A. J.

Dweck

C. S.

Yeager

D. S.

(Eds.) Handbook of competence and motivation: Theory and application (pp. 251−271). The Guilford Press.

31.

Pekrun

Linnenbrink-Garcia

(2012). Academic emotions and student engagement Handbook of research on student engagement (pp. 259−282). Springer. https://doi.org/10.1007/978-1-4614-2018-7_12.

32.

Pekrun

Perry

R. P.

(2014). Control-value theory of achievement emotions. In Pekrun

Linnenbrink-Garcia

(Eds.) International handbook of emotions in education (pp. 120−141). Taylor and Francis.

33.

Peng

Chen

Zhou

(2021). An eye tracking study: positive emotional interface design facilitates learning outcomes in multimedia learning? International Journal of Educational Technology in Higher Education, 18(1), 1−18. https://doi.org/10.1186/s41239-021-00274-x.

34.

Plass

J. L.

Heidig

Hayward

E. O.

Homer

B. D.

(2014). Emotional design in multimedia learning: Effects of shape and color on affect and learning. Learning and Instruction, 29, 128−140. https://doi.org/10.1016/j.learninstruc.2013.02.006.

35.

Plass

J. L.

Homer

B. D.

MacNamara

Ober

Rose

M. C.

Pawar

Hovey

C. M.

Olsen

(2020). Emotional design for digital games for learning: The effect of expression, color, shape, and dimensionality on the affective quality of game characters. Learning and Instruction, 70, 101194. https://doi.org/10.1016/j.learninstruc.2019.01.005.

36.

Plass

J. L.

Kalyuga

(2019). Four ways of considering emotion in cognitive load theory. Educational Psychology Review, 31(2), 339−359. https://doi.org/10.1007/s10648-019-09473-5.

37.

Plass

J. L.

Kaplan

(2016). Emotional design in digital media for learning. In Tettegah

Gartmeier

(Eds.) Emotions, technology, design, and learning (pp. 131−161). Elsevier. https://doi.org/10.1016/b978-0-12-801856-9.00007-4.

38.

Ponce

H. R.

Mayer

R. E.

(2014a). An eye movement analysis of highlighting and graphic organizer study aids for learning from expository text. Computers in Human Behavior, 41, 21−32. https://doi.org/10.1016/j.chb.2014.09.010.

39.

Ponce

H. R.

Mayer

R. E.

(2014b). Qualitatively different cognitive processing during online reading primed by different study activities. Computers in Human Behavior, 30, 121−130. https://doi.org/10.1016/j.chb.2013.07.054.

40.

Riva

Mantovani

Capideville

C. S.

Preziosa

Morganti

Villani

Gaggioli

Botella

Alcañiz

(2007). Affective interactions using virtual reality: The link between presence and emotions. CyberPsychology & Behavior, 10(1), 45−56. https://doi.org/10.1089/cpb.2006.9993.

41.

Russell

J. A.

(1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161−1178. https://doi.org/10.1037/h0077714.

42.

Russell

J. A.

(2003). Core affect and the psychological construction of emotion. Psychological Review, 110(1), 145−172. https://doi.org/10.1037/0033-295x.110.1.145.

43.

Schrader

Kalyuga

Plass

J. L.

(2022). Motivation and affect in multimedia learning. In Mayer

R. E.

Fiorella

(Eds.) The Cambridge handbook of multimedia learning (3rd ed., pp. 121−131). Cambridge University Press.

44.

Stark

Brünken

Park

(2018). Emotional text design in multimedia learning: a mixed-methods study using eye tracking. Computers & Education, 120, 185−196. https://doi.org/10.1016/j.compedu.2018.02.003.

45.

Stárková

Lukavský

Javora

Brom

(2019). Anthropomorphisms in multimedia learning: Attract attention but do not enhance learning? Journal of Computer Assisted Learning, 35(4), 555−568. https://doi.org/10.1111/jcal.12359.

46.

Tettegah

S. Y.

McCreery

M. P.

(Eds.) (2015). Emotions, Technology, and Learning. Academic Press.

47.

E. R.

Plass

J. L.

Hayward

E. O.

Homer

B. D.

(2012). Emotional design in multimedia learning. Journal of Educational Psychology, 104(2), 485−498. https://doi.org/10.1037/a0026609.

48.

Uzun

A. M.

Yıldırım

(2018). Exploring the effect of using different levels of emotional design features in multimedia science learning. Computers & Education, 119, 112−128. https://doi.org/10.1016/j.compedu.2018.01.002.

49.

Wang

Lin

Han

Spector

J. M.

(2020). Impacts of cues on learning: Using eye-tracking technologies to examine the functions and designs of added cues in short instructional videos. Computers in Human Behavior, 107, 106279. https://doi.org/10.1016/j.chb.2020.106279.

50.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: the PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063−1070. https://doi.org/10.1037//0022-3514.54.6.1063.

51.

Wong

R. M.

Adesope

O. O.

(2020). Meta-analysis of emotional designs in multimedia learning: A replication and extension study. Educational Psychology Review, 33(2), 357−385. https://doi.org/10.1007/s10648-020-09545-x.

52.

Zhai

Fang

Dong

Wei

Yuan

Cacciolatti

Yang

(2018). The effects of biofeedback‐based stimulated recall on self‐regulated online learning: A gender and cognitive taxonomy perspective. Journal of Computer Assisted Learning, 34(6), 775−786. https://doi.org/10.1111/jcal.12284.