- Email: [email protected]
Exploring Affective Reaction during User Interaction with Colors and Shapes
Available online at www.sciencedirect.com
ScienceDirect Procedia Manufacturing 3 (2015) 5253 – 5260
6th International Conference on Applied Human Factors and Ergonomics (AHFE 2015) and the Affiliated Conferences, AHFE 2015
Exploring Affective Reaction During User Interaction With Colors And Shapes Tamirat Abegaza*, Edward Dillonb, Juan E. Gilbertb a
b
Clemson University, School of Computing, 100 McAdams Hall, Clemson, SC and 29634, USA University of Florida, Department of Computer & Information Science and Engineering, P.O. Box 116120, Gainesville, FL and 32611, USA
Abstract The interactive structure of software applications involves multiple facets of graphic design and information architecture. A user’s affective reaction can be influenced while interacting with these applications. From a user’s experience (UX) point of view, the affective reaction generated from users in some cases can be considered of an equal or even higher importance than the design and architecture of a software application. When describing affective reaction within the context of user experience, a change in the users’ attitude could also influence a change in their behavior while interacting with certain applications. More importantly, this behavior change could impact the users’ quality of decision-making during interaction. Affective reactions can be invoked and influenced through the incorporation of visual emotional stimuli. Example visual emotional stimuli include colors, facial expressions, movie scenes, and shapes of objects. Amongst these particular stimuli, colors and shapes can be considered of a lower level design than facial expressions and movie scenes. Previous studies have provided mixed findings regarding an individual’s affective reaction when exposed to colors and shapes. However, this article investigates the potential of employing such elements as viable alternatives to higher level stimuli such as facial expression and movie scenes. A study was conducted to explore affective reaction using colors, shapes, and their combinations to serve as visual emotional stimuli. The objective was to understand the possibility of using lower level design alternatives for manipulating users’ affective reactions by comparing colors and shapes in the form of textual properties, abstract polygonal shapes, and the combination of both respectively. The study consisted of thirteen participants who were exposed to these design elements in their variety while interacting with a simple computer application. For measurement and analysis, one-way repeated measure ANOVAs were employed to determine the effectiveness of these design elements and their mood induction capabilities during user interaction. The results from this study indicated that the presence of colors, shapes, and their combinations were able to invoke certain affective reactions without the participants even noticing their presence. More specifically, this study revealed that certain colors and shapes generated a more positive affective reaction while others induced either a more neutral or less positive emotion from the participants. This study also demonstrated that these particular design elements have high visceral effects that could be used as potential alternatives to more traditional and higher level visual emotional stimuli in order to influence a user’s affective reaction. © Elsevier B.V. This is an open access article under the CC BY-NC-ND license © 2015 2015The TheAuthors. Authors.Published Publishedbyby Elsevier B.V. (http://creativecommons.org/licenses/by-nc-nd/4.0/). under responsibility of AHFE Conference. Peer-review Peer-review under responsibility of AHFE Conference Keywords: Visual Stimuli; Mood Induction; Emotional Design, Affective Reaction; *Corresponding author. Tel.: 336-587-3538; E-mail address: [email protected];
2351-9789 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of AHFE Conference doi:10.1016/j.promfg.2015.07.602
5254
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
1. Introduction Affect matters. It can be influenced by the graphic design, information architecture, or direct interaction with software applications. According to Weiss and Cropanzano, affect is defined as “an unconscious reaction to stimuli before any cognitive appraisal of an event occurs” [1]. It has been found that one’s affective state, whether positive or negative, can play an influential part on one’s cognitive behavior with software applications [1-4]. Furthermore, people’s judgments and decision-making prowess can be influenced by their current affective (or emotional) state. Bower [3] noted that emotion can influence certain judgment(s) due to the biased retrieval of related information from one’s memory. Clore and Jeffrey [4] suggested that emotion can impact the content and style of one’s thoughts. When it comes to particular emotions, Forgas [5] found that individuals with positive moods tend to make faster decisions while exerting a superficial focus on important attributes. Moreover, Loewenstin and Lerner [6] emphasized that positive emotion triggers a belief that all is well, while negative emotion causes one to place a heavier concentration towards the situation of concern. Averill [7] noted that negative emotion is linked to vigilance, while positive emotion is associated with trust. One’s affective (or emotional) state can also be induced by certain environmental stimuli. For instance, smiling facial expressions, symmetrical objects, pleasant odors, and harmonious music/sounds are example stimuli that could potentially induce a positive affective state, whereas angry facial expressions, pungent odors, and discordant sounds could induce a negative affective state [8-11]. From a research standpoint, visual stimuli (ex. facial expression, color), auditory stimuli (vocal expression), and/or the combinations of both (movie scene) have been employed by researchers to serve as catalyst for mood induction [8-11]. From a UX perspective, it may be possible to incorporate stimuli into software application designs in order to invoke certain affective reactions from users. Ideally, the users would exhibit either positive or negative emotion that influences a change in attitude, which is followed by a change in their behavior during interaction. This behavior change could then impact the quality of the users’ decision-making abilities. This article discusses a study that investigated the effect of lower level visual stimuli in the form of colors (ex. background and foreground), shapes (ex. rounded, angular, and mixed), and their combinations on participants and their affective states. The objective of this study was to determine whether the presence of these lower level stimuli could induce certain emotions from these participants upon exposure. The remaining sections of this article are composed as follows: the Related Work section discusses prior research regarding the impacts of lower level stimuli such as colors and shapes on individuals; the Study section talks about the actual study that was conducted to determine the different affective reactions exhibited by a group of participants who were exposed to a variety of colors, shapes, and their combination as stimuli; the Results section provides summarized data and statistical findings generated from the conducted study; a discussion is provided about these results in the Discussion section, while the Conclusion section reveal the conclusive findings from this study along with potential confounding factors and future work. 2. Related Work 2.1 Colors Prior research has noted that colors with shorter wavelengths tend to be more pleasant to individuals than ones with a longer wavelength [12-14]. According to Elliot and Markus’ model for color and psychological functionalities [15], color could be explained using five core principles. First, the meaning of color is derived from innate and learned behavior. This particular principle is based on Mollon’s theory that explains the existence of learned associations with innate response to color stimuli [12]. Second, positive color stimuli produce approach behavior, whereas negative color stimuli invoke avoidance behavior. Third, the influential impact of colors can be linked to one’s unconscious behavior. Fourth, color evokes varied feelings for different situations. Finally, colors can carry particular contextual meanings while present in certain situations. The colors red, blue, and green tend to be studied more frequently than others [12-18]. In particular, research conducted on the color red has generated various findings from researchers [14-15, 19-23]. Table 1 provides a summarized list of these findings.
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
5255
Table 1. Various Findings from Studying the Color Red Authors [citation]
Purpose of using Color
Findings
Stone & English, 1998 [19]
Room Wall Color
Improved performance motor task
Hill & Barton, 2005 [20]
Clothing
Improved performance for those who wear red
Elliot & Maier, 2007[15]
Background Color
Undermined intellectual performance
Mehta & Rui, 2009 [14]
Background Color
Improved performance on detailedoriented task
Gnambs et al., 2010 [21]
Progress Bar, Buttons
Impaired test performance
Thorstenson, 2012 [22]
Pen Ink
Improved performance on detailedoriented tasks (marked more errors when correcting essays using red ink)
2.2 Shapes Shapes can contain angular, rounded, or mixed edges. Research indicates that the shapes of visual objects are strongly associated with emotion [24]. Studies conducted on shapes tend to report that rounder edges are associated with positive affect, while angular edges elicit negative affect [24-26]. Angular stimuli have also been found to trigger threats [27]. For instance, an angular edge on sharp objects, like knives, could be inherently associated with some level of harm or danger. However, objects with rounder shapes, like the face of babies, could impose the exact opposite association [26]. It has also been found that the shapes of visual stimuli can still be associated with emotion regardless of an object’s semantic meaning or purpose. For instance, studies have shown that emotionally neutral objects with curved features tend to invoke positive affect, while related objects with angular/sharper features triggered negative affect [26-29]. In particular, Bar and Meta [26] studied the effect of shapes using watches, guitars, and letters of the alphabet. Their findings revealed that the objects containing curved shapes increased the participants’ level of positive affect. 2.3 Discussion Findings from the related work have shown that colors and shapes respectively influence certain emotions from individuals upon exposure. For instance, a color’s wavelength can serve as a potential factor for influencing one’s mood, behavior, or feelings, while a shape’s edge can impose a similar outcome. For the study discussed in this article, abstract design stimuli was used to investigate the emotion induction capabilities of objects while using varying colors and curvatures. Specifically, the design contained colors possessing varying wavelengths along with shapes containing both rounded and angular edges. The next section discusses the experimental procedure used to conduct this study. 3. Study 3.1 Method For this study, three emotional design conditions were explored (see Figures 1a-c): foreground color as textual property/word cloud (Figure 1a), shape as polygonal property (Figure 1b), and color/shape combination as background color and polygonal property (Figure 1c). The textual property word cloud presents three color conditions: blue, black, and red foreground colors to potentially induce high, neutral, and low levels of positive affect respectively. Similarly, the polygonal properties illustrate rounded, mixed, and angular shapes that were used to potentially induce high, neutral, and low levels of positive affect respectively. Finally, the background color/polygonal properties show the combinations of round/blue, mixed/gray, and angular/red design elements to potentially induce high, neutral, and low levels of positive affect respectively.
5256
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
Textual Properties/Word Clouds
Polygonal Properties
Background color/Polygonal Properties
High Level
High Level
High Level
Neutral
Neutral
Neutral
Low Level
Low Level
Low Level
(a)
(b)
(c)
Fig. 1a-c. Design Element Conditions
3.2 Participants There were thirteen participants recruited for this study. Seven of the participants were female. The participants were all graduate students at Clemson University who aged between 20 and 35. Table 2 provides further details regarding demographic representation of the participant pool. The Self-Assessment-Manikin (SAM) [30] model was used to rate the emotional scale of the users during this study (see Figure 2). 3.3 Study Procedure To begin the study, each participant was asked to sit in a chair facing the designated computer and its screen. From the screen, appeared a simple computer application that prompted the participant to initially rate his or her current emotional state. This screen stayed active until the participant chose his or her rating. Afterwards, the application showed a blank window that lasted for two seconds. After the two-second lapse, the application prompted the participant to rate his or her current emotional state. The application then revealed a one-second long stimulus that included one of the emotional design elements. The participant was then prompted to rate his or her current emotional state. This was followed by a two-second inter-stimulus interval that presented a blank window in order to neutralize the impact of the prior emotional experience. This procedural pattern continued until the participant was exposed to every design element (colors, shapes, and their combinations). Figure 2 provides an illustrated depiction of the procedure for this study. Table 2. Demographic Data on Participants (N=13) Percentage
Number of Participants
Education College Degree Graduate Degree
61.5% 38.5%
8 5
Race African American Caucasian Hispanic Asian
53.9% 23.1% 7.7% 15.4%
7 3 1 2
Age 21 and under 22-35
7.7% 92.3%
1 12
Gender Female Male
53.9% 46.1%
7 6
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
5257
Fig. 2. The Process of Mood Induction Rating for the Study
3.3 Study & Analysis Design To measure the emotional ratings more precisely, means scores were used over median. Furthermore, the order of design element exposure was determined using an incomplete/partial counterbalancing Latin square model. This model was used in order to reduce any learning effect, biases, and tiredness potentially exhibited from the participants. Each design element was tested nine times (three trials for each of the conditions). The participants’ mood induction ratings were based on a 5-point Likert scale (1=strongly negative, 2=negative, 3=neutral, 4=positive, and 5=strongly positive). The mean of the three trials per condition was used to obtain a single value that represented the mood ratings of the participants. For statistical analysis, one-way ANOVAs were used to determine any statistical significance of color, shape, or color/shape combination design elements. 4. Results 4.1 Textual Properties/Word Clouds The overall mood ratings (using means) for the designs that employed textual property colors in the form of word clouds (WC: WC-Blue, WC-Black, and WC-Red) as visual stimuli were 3.54, 3.13, and 2.49 respectively. This indicated that the participants’ exposure to the blue foreground color within the text induced a more positive mood, whereas the red foreground color generated a less positive mood. The black foreground color was found to induce a more neutral mood from the participants in comparison to the blue and red colors. Figure 3 provides a rating plot for the mean mood ratings for each condition.
Fig. 3. Average mood induction rating plot for color based textual property
When conducting statistical analysis on the means for the three conditions (WC-Blue, WC-Black and WC-Red), a one-way ANOVA revealed a statistical significance: F (2, 24)=13.96, p<0.0001. Bonferroni tests were used to
5258
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
make post-hoc comparisons between the three conditions. These tests revealed a statistical significance for two of the three paired comparisons: WC-Blue vs. WC-Red (p<0.0001) and WC-Black vs. WC-Red (p=0.047). These results suggest that the employment of color as part of a design element has a potential mood induction capability. In particular, blue color text induced a more positive emotion from the participants while the red color text induced a less positive emotion from the participants. The black color text induced a more neutral emotion from the participants. 4.2 Shape Properties The overall mood ratings (using means) for the designs that employed objects in the form of rounded, mixed, and angular shapes were 3.57, 2.95, and 2.51 respectively. This indicated that the participants’ exposure to the rounded shapes induced a more positive mood, whereas the angular shapes generated a less positive mood. The mixed shapes were found to induce a more neutral mood in comparison to the rounded and angular shapes. Figure 4 provides a rating plot for the mean mood ratings for each condition. When conducting statistical analysis on the means for the three conditions (Round, Mixed, and Angular), a oneway ANOVA revealed a statistical significance: F (1.15, 13.8)=12.49, p=0.003. Bonferroni tests were used to make post-hoc comparisons between the three conditions. These tests revealed a statistical significance for two of the three pair comparisons: Round vs. Angular (p<0.01) and Round vs. Mixed (p=0.003). These results suggest that the employment of shapes as part of a design element has a potential mood induction capability. In particular, the rounded shapes induced a more positive emotion whereas the angular shapes induced a less positive emotion from the participants. The mixed shapes induced a more neutral emotion from the participants.
Fig. 4. Average mood induction rating plot for shape conditions: Round, Mixed, and Angular
4.3 Color/Shape Combination The overall mood ratings (using means) for the designs that employed objects that used combinations of certain shapes and colors, in particular RoundBlue, MixedGray, and AngularRed, were 3.82, 3.03, and 2.49 respectively. This indicated that the participants’ exposure to the RoundBlue stimuli induced a more positive mood, whereas the AngularRed stimuli generated a less positive mood. The MixedGray stimuli were found to induce a more neutral mood from the participants in comparison to the RoundBlue and AngularRed stimuli. Figure 5 provides a rating plot for the mean mood ratings for each condition.
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
5259
Fig. 5. Average mood induction rating plot for color/shape conditions: RoundBlue, MixedGray, and AngularRed
When conducting statistical analysis on the means for the three conditions (RoundBlue, MixedGray, AngularRed), a one-way ANOVA revealed a statistical significance: F (2, 24)=14.96, p<0.0001. Bonferroni tests were used to make post-hoc comparisons between the three conditions. These tests revealed a statistical significance for two of the three pair comparisons: RoundBlue vs. AngularRed (p=0.002) and RoundBlue vs. MixedGray (p=0.002). These results suggest that the blue color and round shape combination induced a more positive emotion from the participants, while the red color and angular shape combination induced a less positive emotion. The gray color and mixed shape combination induced a more neutral emotion from the participants. 5. Discussion The results revealed that the presence of blue, black, and red colors tended to induce more positive, neutral, and less positive emotions respectively from the participants. Even though previous studies showed different outcomes when studying the color red (as mentioned in the Related Work section), this particular study revealed that the inclusion of this color as part of a design element could potentially result in decreased positive emotion. In regards to shapes, the results showed that the employment of angular and rounder shapes could induce less positive and more positive emotion respectively from the participants. Overall, the results showed that the emotional design elements used in this study have emotion induction possibilities. Furthermore, these particular design elements could be used as alternate mood induction stimuli to facial expression, voice, or multimedia that are also used to induce affective reactions. Hence, this study demonstrated the possibility that low level emotional design elements such colors and shapes have high visceral effects that could be used to induce certain emotional states without the users’ awareness. 6. Conclusion (Threats to Validity/Future Work) The goal of this article was to discuss the employment of lower level design elements (such as colors and shapes) and their potential to induce certain affective states from users during exposure. The study presented in this article demonstrated that such elements could be used as viable alternatives to higher level stimuli like facial expression and movie scenes due to their high visceral effects. In particular, the study revealed that certain colors, shapes, and their combinations potentially induced a change in the users affective state (whether more positive, neutral, or less positive) upon exposure. Even though the intent of this study was to determine whether certain design elements could potentially induce affective reactions from users, there were some potential confounding factors. One, the emotional ratings scale used during this study was based on self-related emotion, which could be influenced by various factors besides the imposed stimuli. Another potential factor was the limited representation of participants involved in this study. Employing this study on participant samples that are more diverse and greater in number would not only improve the representation of the general population but also the accuracy of these findings. For future work, these elements will be used to examine their relative impact on user decision making while using a software application to conduct information search and retrieval. Areas of focus will be placed on the users’ search thoroughness and overall satisfaction during interaction.
5260
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
Acknowledgments A special thanks to the Human-Centered Computing lab at Clemson University for providing the platform to conduct this study. The authors would also like to acknowledge and thank the participants who were involved in this study. References [1]
H. Weiss, R. Cropanzano, Affective Events Theory: A theoretical discussion of the structure, causes and consequences of affective experiences at work, in: B.M. Staw, L.L. Cummings (Eds.), Research in organizational behavior: An annual series of analytical essays and critical reviews. 18 (1996) 1-74.
[2]
J. Basch, C. D. Fisher, Affective events-emotions matrix: A classification of work events and associated emotions, School of Business Discussion Papers #65 (1998) 1-22.
[3]
B. Bower, When good moods go decisively bad Positive feelings can lead to less than ideal choices in seniors, A+ A.10 (2012).
[4]
G. L. Clore, J. R. Huntsinger, How emotions inform judgment and regulate thought, Trends in cognitive sciences. 11.9 (2007) 393-399.
[5]
J.P. Forgas, Affective influences on partner choice: Role of mood in social decisions, Journal of Personality and Social Psychology. 61.5 (1991) 708-720.
[6]
G. Loewenstein, J. S. Lerner, The role of affect in decision making, Handbook of affective science. (2003) 619-642.
[7]
J.R. Averill, Studies on anger and aggression: Implications for theories of emotion, American Psychologist. 38.11 (1983) 1145-1160.
[8]
L.L. Carstensen, B. Turan, S. Scheibe, N. Ram, H. Ersner-Hershfield, G. R. Samanez-Larkin, K. P. Brooks, J. R. Nesselroade. Emotional experience improves with age: evidence based on over 10 years of experience sampling, Psychology and aging. 26.1 (2011) 21-33.
[9]
D. McDuff, R. El Kaliouby, D. Demirdjian, R. Picard, Predicting online media effectiveness based on smile responses gathered over the internet, Automatic Face and Gesture Recognition (FG). 10th IEEE International Conference and Workshops. (2013) 1-7.
[10] R. D. Lane, P. ML. Chua, R. J. Dolan, Common effects of emotional valence, arousal and attention on neural activation during visual processing of pictures, Neuropsychologia. 37.9 (1999) 989-997. [11] R. J. Dolan, Emotion, cognition, and behaviour, Science. 298.5596 (2002) 1191-1194. [12] J.D. Mollon, "“Tho'she kneel'd in that place where they grew…” The uses and origins of primate color vision." Journal of Experimental Biology. 146.1 (1989) 21-38. [13] T. Hatta, H. Yoshida, A. Kawakami, M. Okamoto, Color of computer display frame in work performance, mood, and physiological response, Perceptual and motor skills. 94.1 (2002) 39-46. [14] R. Mehta, R. J. Zhu, Blue or red? Exploring the effect of color on cognitive task performances, Science. 323.5918 (2009) 1226-1229. [15] A.J. Elliot, M. A. Maier, Color and psychological functioning, Current Directions in Psychological Science. 16.5 (2007) 250-254. [16] K.W. Jacobs, F. E. Hustmyer Jr, Effects of four psychological primary colors on GSR, heart rate and respiration rate, Perceptual and motor skills. 38.3 (1974) 763-766. [17] N. P. Juslin, K. R. Scherer, Vocal expression of affect, The new handbook of methods in nonverbal behavior research. (2005) 65-135. [18] C. Caffi, R.W. Janney, Toward a pragmatics of emotive communication, Journal of Pragmatics. 22(3–4), (1994) 325–373. [19] N.J. Stone, A.J. English, Task type, posters, and workspace color on mood, satisfaction, and performance, Journal of Environmental Psychology. 18 (1998) 175–185. [20] R.A. Hill, R. A. Barton, Psychology: red enhances human performance in contests, Nature. 435.7040 (2005) 293-293. [21] T. Gnambs, M. Appel, B. Batinic, Color red in web-based knowledge testing, Computers in Human Behavior. 26.6 (2010) 1625-1631. [22] C.A. Thorstenson, Effects of Color Perception and Enacted Avoidance Behavior on Intellectual Task Performance in an Achievement Context, Diss. Appalachian State University, 2012. [23] R. Bagchi, A. Cheema, The effect of red background color on willingness-to-pay: The moderating role of selling mechanism, Journal of Consumer Research. 39.5 (2013) 947-960. [24] M. Bar, A cortical mechanism for triggering top-down facilitation in visual object recognition, Journal of Cognitive Neuroscience. 15.4 (2003) 600-609. [25] M. Bar, M.Neta, H. Linz, Very first impressions, Emotion. 6.2 (2006) 269-278. [26] M. Bar, M. Neta, Humans prefer curved visual objects, Psychological science. 17.8 (2006) 645-648. [27] L.A. Zebrowitz, Reading faces: Window to the soul?, New Directions in Social Psychology, Westview Press, Boulder, CO, 1997. [28] J. Aronoff, B. A. Woike, L. M. Hyman, Which are the stimuli in facial displays of anger and happiness? Configurational bases of emotion recognition, Journal of Personality and Social Psychology. 62.6 (1992) 1050-1066. [29] M. Bar, The proactive brain: using analogies and associations to generate predictions, Trends in cognitive sciences. 11.7 (2007) 280-289. [30] M.M. Bradley, P. J. Lang, Measuring emotion: the self-assessment manikin and the semantic differential, Journal of behavior therapy and experimental psychiatry. 25.1 (1994): 49-59.
ScienceDirect Procedia Manufacturing 3 (2015) 5253 – 5260
6th International Conference on Applied Human Factors and Ergonomics (AHFE 2015) and the Affiliated Conferences, AHFE 2015
Exploring Affective Reaction During User Interaction With Colors And Shapes Tamirat Abegaza*, Edward Dillonb, Juan E. Gilbertb a
b
Clemson University, School of Computing, 100 McAdams Hall, Clemson, SC and 29634, USA University of Florida, Department of Computer & Information Science and Engineering, P.O. Box 116120, Gainesville, FL and 32611, USA
Abstract The interactive structure of software applications involves multiple facets of graphic design and information architecture. A user’s affective reaction can be influenced while interacting with these applications. From a user’s experience (UX) point of view, the affective reaction generated from users in some cases can be considered of an equal or even higher importance than the design and architecture of a software application. When describing affective reaction within the context of user experience, a change in the users’ attitude could also influence a change in their behavior while interacting with certain applications. More importantly, this behavior change could impact the users’ quality of decision-making during interaction. Affective reactions can be invoked and influenced through the incorporation of visual emotional stimuli. Example visual emotional stimuli include colors, facial expressions, movie scenes, and shapes of objects. Amongst these particular stimuli, colors and shapes can be considered of a lower level design than facial expressions and movie scenes. Previous studies have provided mixed findings regarding an individual’s affective reaction when exposed to colors and shapes. However, this article investigates the potential of employing such elements as viable alternatives to higher level stimuli such as facial expression and movie scenes. A study was conducted to explore affective reaction using colors, shapes, and their combinations to serve as visual emotional stimuli. The objective was to understand the possibility of using lower level design alternatives for manipulating users’ affective reactions by comparing colors and shapes in the form of textual properties, abstract polygonal shapes, and the combination of both respectively. The study consisted of thirteen participants who were exposed to these design elements in their variety while interacting with a simple computer application. For measurement and analysis, one-way repeated measure ANOVAs were employed to determine the effectiveness of these design elements and their mood induction capabilities during user interaction. The results from this study indicated that the presence of colors, shapes, and their combinations were able to invoke certain affective reactions without the participants even noticing their presence. More specifically, this study revealed that certain colors and shapes generated a more positive affective reaction while others induced either a more neutral or less positive emotion from the participants. This study also demonstrated that these particular design elements have high visceral effects that could be used as potential alternatives to more traditional and higher level visual emotional stimuli in order to influence a user’s affective reaction. © Elsevier B.V. This is an open access article under the CC BY-NC-ND license © 2015 2015The TheAuthors. Authors.Published Publishedbyby Elsevier B.V. (http://creativecommons.org/licenses/by-nc-nd/4.0/). under responsibility of AHFE Conference. Peer-review Peer-review under responsibility of AHFE Conference Keywords: Visual Stimuli; Mood Induction; Emotional Design, Affective Reaction; *Corresponding author. Tel.: 336-587-3538; E-mail address: [email protected];
2351-9789 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of AHFE Conference doi:10.1016/j.promfg.2015.07.602
5254
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
1. Introduction Affect matters. It can be influenced by the graphic design, information architecture, or direct interaction with software applications. According to Weiss and Cropanzano, affect is defined as “an unconscious reaction to stimuli before any cognitive appraisal of an event occurs” [1]. It has been found that one’s affective state, whether positive or negative, can play an influential part on one’s cognitive behavior with software applications [1-4]. Furthermore, people’s judgments and decision-making prowess can be influenced by their current affective (or emotional) state. Bower [3] noted that emotion can influence certain judgment(s) due to the biased retrieval of related information from one’s memory. Clore and Jeffrey [4] suggested that emotion can impact the content and style of one’s thoughts. When it comes to particular emotions, Forgas [5] found that individuals with positive moods tend to make faster decisions while exerting a superficial focus on important attributes. Moreover, Loewenstin and Lerner [6] emphasized that positive emotion triggers a belief that all is well, while negative emotion causes one to place a heavier concentration towards the situation of concern. Averill [7] noted that negative emotion is linked to vigilance, while positive emotion is associated with trust. One’s affective (or emotional) state can also be induced by certain environmental stimuli. For instance, smiling facial expressions, symmetrical objects, pleasant odors, and harmonious music/sounds are example stimuli that could potentially induce a positive affective state, whereas angry facial expressions, pungent odors, and discordant sounds could induce a negative affective state [8-11]. From a research standpoint, visual stimuli (ex. facial expression, color), auditory stimuli (vocal expression), and/or the combinations of both (movie scene) have been employed by researchers to serve as catalyst for mood induction [8-11]. From a UX perspective, it may be possible to incorporate stimuli into software application designs in order to invoke certain affective reactions from users. Ideally, the users would exhibit either positive or negative emotion that influences a change in attitude, which is followed by a change in their behavior during interaction. This behavior change could then impact the quality of the users’ decision-making abilities. This article discusses a study that investigated the effect of lower level visual stimuli in the form of colors (ex. background and foreground), shapes (ex. rounded, angular, and mixed), and their combinations on participants and their affective states. The objective of this study was to determine whether the presence of these lower level stimuli could induce certain emotions from these participants upon exposure. The remaining sections of this article are composed as follows: the Related Work section discusses prior research regarding the impacts of lower level stimuli such as colors and shapes on individuals; the Study section talks about the actual study that was conducted to determine the different affective reactions exhibited by a group of participants who were exposed to a variety of colors, shapes, and their combination as stimuli; the Results section provides summarized data and statistical findings generated from the conducted study; a discussion is provided about these results in the Discussion section, while the Conclusion section reveal the conclusive findings from this study along with potential confounding factors and future work. 2. Related Work 2.1 Colors Prior research has noted that colors with shorter wavelengths tend to be more pleasant to individuals than ones with a longer wavelength [12-14]. According to Elliot and Markus’ model for color and psychological functionalities [15], color could be explained using five core principles. First, the meaning of color is derived from innate and learned behavior. This particular principle is based on Mollon’s theory that explains the existence of learned associations with innate response to color stimuli [12]. Second, positive color stimuli produce approach behavior, whereas negative color stimuli invoke avoidance behavior. Third, the influential impact of colors can be linked to one’s unconscious behavior. Fourth, color evokes varied feelings for different situations. Finally, colors can carry particular contextual meanings while present in certain situations. The colors red, blue, and green tend to be studied more frequently than others [12-18]. In particular, research conducted on the color red has generated various findings from researchers [14-15, 19-23]. Table 1 provides a summarized list of these findings.
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
5255
Table 1. Various Findings from Studying the Color Red Authors [citation]
Purpose of using Color
Findings
Stone & English, 1998 [19]
Room Wall Color
Improved performance motor task
Hill & Barton, 2005 [20]
Clothing
Improved performance for those who wear red
Elliot & Maier, 2007[15]
Background Color
Undermined intellectual performance
Mehta & Rui, 2009 [14]
Background Color
Improved performance on detailedoriented task
Gnambs et al., 2010 [21]
Progress Bar, Buttons
Impaired test performance
Thorstenson, 2012 [22]
Pen Ink
Improved performance on detailedoriented tasks (marked more errors when correcting essays using red ink)
2.2 Shapes Shapes can contain angular, rounded, or mixed edges. Research indicates that the shapes of visual objects are strongly associated with emotion [24]. Studies conducted on shapes tend to report that rounder edges are associated with positive affect, while angular edges elicit negative affect [24-26]. Angular stimuli have also been found to trigger threats [27]. For instance, an angular edge on sharp objects, like knives, could be inherently associated with some level of harm or danger. However, objects with rounder shapes, like the face of babies, could impose the exact opposite association [26]. It has also been found that the shapes of visual stimuli can still be associated with emotion regardless of an object’s semantic meaning or purpose. For instance, studies have shown that emotionally neutral objects with curved features tend to invoke positive affect, while related objects with angular/sharper features triggered negative affect [26-29]. In particular, Bar and Meta [26] studied the effect of shapes using watches, guitars, and letters of the alphabet. Their findings revealed that the objects containing curved shapes increased the participants’ level of positive affect. 2.3 Discussion Findings from the related work have shown that colors and shapes respectively influence certain emotions from individuals upon exposure. For instance, a color’s wavelength can serve as a potential factor for influencing one’s mood, behavior, or feelings, while a shape’s edge can impose a similar outcome. For the study discussed in this article, abstract design stimuli was used to investigate the emotion induction capabilities of objects while using varying colors and curvatures. Specifically, the design contained colors possessing varying wavelengths along with shapes containing both rounded and angular edges. The next section discusses the experimental procedure used to conduct this study. 3. Study 3.1 Method For this study, three emotional design conditions were explored (see Figures 1a-c): foreground color as textual property/word cloud (Figure 1a), shape as polygonal property (Figure 1b), and color/shape combination as background color and polygonal property (Figure 1c). The textual property word cloud presents three color conditions: blue, black, and red foreground colors to potentially induce high, neutral, and low levels of positive affect respectively. Similarly, the polygonal properties illustrate rounded, mixed, and angular shapes that were used to potentially induce high, neutral, and low levels of positive affect respectively. Finally, the background color/polygonal properties show the combinations of round/blue, mixed/gray, and angular/red design elements to potentially induce high, neutral, and low levels of positive affect respectively.
5256
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
Textual Properties/Word Clouds
Polygonal Properties
Background color/Polygonal Properties
High Level
High Level
High Level
Neutral
Neutral
Neutral
Low Level
Low Level
Low Level
(a)
(b)
(c)
Fig. 1a-c. Design Element Conditions
3.2 Participants There were thirteen participants recruited for this study. Seven of the participants were female. The participants were all graduate students at Clemson University who aged between 20 and 35. Table 2 provides further details regarding demographic representation of the participant pool. The Self-Assessment-Manikin (SAM) [30] model was used to rate the emotional scale of the users during this study (see Figure 2). 3.3 Study Procedure To begin the study, each participant was asked to sit in a chair facing the designated computer and its screen. From the screen, appeared a simple computer application that prompted the participant to initially rate his or her current emotional state. This screen stayed active until the participant chose his or her rating. Afterwards, the application showed a blank window that lasted for two seconds. After the two-second lapse, the application prompted the participant to rate his or her current emotional state. The application then revealed a one-second long stimulus that included one of the emotional design elements. The participant was then prompted to rate his or her current emotional state. This was followed by a two-second inter-stimulus interval that presented a blank window in order to neutralize the impact of the prior emotional experience. This procedural pattern continued until the participant was exposed to every design element (colors, shapes, and their combinations). Figure 2 provides an illustrated depiction of the procedure for this study. Table 2. Demographic Data on Participants (N=13) Percentage
Number of Participants
Education College Degree Graduate Degree
61.5% 38.5%
8 5
Race African American Caucasian Hispanic Asian
53.9% 23.1% 7.7% 15.4%
7 3 1 2
Age 21 and under 22-35
7.7% 92.3%
1 12
Gender Female Male
53.9% 46.1%
7 6
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
5257
Fig. 2. The Process of Mood Induction Rating for the Study
3.3 Study & Analysis Design To measure the emotional ratings more precisely, means scores were used over median. Furthermore, the order of design element exposure was determined using an incomplete/partial counterbalancing Latin square model. This model was used in order to reduce any learning effect, biases, and tiredness potentially exhibited from the participants. Each design element was tested nine times (three trials for each of the conditions). The participants’ mood induction ratings were based on a 5-point Likert scale (1=strongly negative, 2=negative, 3=neutral, 4=positive, and 5=strongly positive). The mean of the three trials per condition was used to obtain a single value that represented the mood ratings of the participants. For statistical analysis, one-way ANOVAs were used to determine any statistical significance of color, shape, or color/shape combination design elements. 4. Results 4.1 Textual Properties/Word Clouds The overall mood ratings (using means) for the designs that employed textual property colors in the form of word clouds (WC: WC-Blue, WC-Black, and WC-Red) as visual stimuli were 3.54, 3.13, and 2.49 respectively. This indicated that the participants’ exposure to the blue foreground color within the text induced a more positive mood, whereas the red foreground color generated a less positive mood. The black foreground color was found to induce a more neutral mood from the participants in comparison to the blue and red colors. Figure 3 provides a rating plot for the mean mood ratings for each condition.
Fig. 3. Average mood induction rating plot for color based textual property
When conducting statistical analysis on the means for the three conditions (WC-Blue, WC-Black and WC-Red), a one-way ANOVA revealed a statistical significance: F (2, 24)=13.96, p<0.0001. Bonferroni tests were used to
5258
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
make post-hoc comparisons between the three conditions. These tests revealed a statistical significance for two of the three paired comparisons: WC-Blue vs. WC-Red (p<0.0001) and WC-Black vs. WC-Red (p=0.047). These results suggest that the employment of color as part of a design element has a potential mood induction capability. In particular, blue color text induced a more positive emotion from the participants while the red color text induced a less positive emotion from the participants. The black color text induced a more neutral emotion from the participants. 4.2 Shape Properties The overall mood ratings (using means) for the designs that employed objects in the form of rounded, mixed, and angular shapes were 3.57, 2.95, and 2.51 respectively. This indicated that the participants’ exposure to the rounded shapes induced a more positive mood, whereas the angular shapes generated a less positive mood. The mixed shapes were found to induce a more neutral mood in comparison to the rounded and angular shapes. Figure 4 provides a rating plot for the mean mood ratings for each condition. When conducting statistical analysis on the means for the three conditions (Round, Mixed, and Angular), a oneway ANOVA revealed a statistical significance: F (1.15, 13.8)=12.49, p=0.003. Bonferroni tests were used to make post-hoc comparisons between the three conditions. These tests revealed a statistical significance for two of the three pair comparisons: Round vs. Angular (p<0.01) and Round vs. Mixed (p=0.003). These results suggest that the employment of shapes as part of a design element has a potential mood induction capability. In particular, the rounded shapes induced a more positive emotion whereas the angular shapes induced a less positive emotion from the participants. The mixed shapes induced a more neutral emotion from the participants.
Fig. 4. Average mood induction rating plot for shape conditions: Round, Mixed, and Angular
4.3 Color/Shape Combination The overall mood ratings (using means) for the designs that employed objects that used combinations of certain shapes and colors, in particular RoundBlue, MixedGray, and AngularRed, were 3.82, 3.03, and 2.49 respectively. This indicated that the participants’ exposure to the RoundBlue stimuli induced a more positive mood, whereas the AngularRed stimuli generated a less positive mood. The MixedGray stimuli were found to induce a more neutral mood from the participants in comparison to the RoundBlue and AngularRed stimuli. Figure 5 provides a rating plot for the mean mood ratings for each condition.
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
5259
Fig. 5. Average mood induction rating plot for color/shape conditions: RoundBlue, MixedGray, and AngularRed
When conducting statistical analysis on the means for the three conditions (RoundBlue, MixedGray, AngularRed), a one-way ANOVA revealed a statistical significance: F (2, 24)=14.96, p<0.0001. Bonferroni tests were used to make post-hoc comparisons between the three conditions. These tests revealed a statistical significance for two of the three pair comparisons: RoundBlue vs. AngularRed (p=0.002) and RoundBlue vs. MixedGray (p=0.002). These results suggest that the blue color and round shape combination induced a more positive emotion from the participants, while the red color and angular shape combination induced a less positive emotion. The gray color and mixed shape combination induced a more neutral emotion from the participants. 5. Discussion The results revealed that the presence of blue, black, and red colors tended to induce more positive, neutral, and less positive emotions respectively from the participants. Even though previous studies showed different outcomes when studying the color red (as mentioned in the Related Work section), this particular study revealed that the inclusion of this color as part of a design element could potentially result in decreased positive emotion. In regards to shapes, the results showed that the employment of angular and rounder shapes could induce less positive and more positive emotion respectively from the participants. Overall, the results showed that the emotional design elements used in this study have emotion induction possibilities. Furthermore, these particular design elements could be used as alternate mood induction stimuli to facial expression, voice, or multimedia that are also used to induce affective reactions. Hence, this study demonstrated the possibility that low level emotional design elements such colors and shapes have high visceral effects that could be used to induce certain emotional states without the users’ awareness. 6. Conclusion (Threats to Validity/Future Work) The goal of this article was to discuss the employment of lower level design elements (such as colors and shapes) and their potential to induce certain affective states from users during exposure. The study presented in this article demonstrated that such elements could be used as viable alternatives to higher level stimuli like facial expression and movie scenes due to their high visceral effects. In particular, the study revealed that certain colors, shapes, and their combinations potentially induced a change in the users affective state (whether more positive, neutral, or less positive) upon exposure. Even though the intent of this study was to determine whether certain design elements could potentially induce affective reactions from users, there were some potential confounding factors. One, the emotional ratings scale used during this study was based on self-related emotion, which could be influenced by various factors besides the imposed stimuli. Another potential factor was the limited representation of participants involved in this study. Employing this study on participant samples that are more diverse and greater in number would not only improve the representation of the general population but also the accuracy of these findings. For future work, these elements will be used to examine their relative impact on user decision making while using a software application to conduct information search and retrieval. Areas of focus will be placed on the users’ search thoroughness and overall satisfaction during interaction.
5260
Tamirat Abegaz et al. / Procedia Manufacturing 3 (2015) 5253 – 5260
Acknowledgments A special thanks to the Human-Centered Computing lab at Clemson University for providing the platform to conduct this study. The authors would also like to acknowledge and thank the participants who were involved in this study. References [1]
H. Weiss, R. Cropanzano, Affective Events Theory: A theoretical discussion of the structure, causes and consequences of affective experiences at work, in: B.M. Staw, L.L. Cummings (Eds.), Research in organizational behavior: An annual series of analytical essays and critical reviews. 18 (1996) 1-74.
[2]
J. Basch, C. D. Fisher, Affective events-emotions matrix: A classification of work events and associated emotions, School of Business Discussion Papers #65 (1998) 1-22.
[3]
B. Bower, When good moods go decisively bad Positive feelings can lead to less than ideal choices in seniors, A+ A.10 (2012).
[4]
G. L. Clore, J. R. Huntsinger, How emotions inform judgment and regulate thought, Trends in cognitive sciences. 11.9 (2007) 393-399.
[5]
J.P. Forgas, Affective influences on partner choice: Role of mood in social decisions, Journal of Personality and Social Psychology. 61.5 (1991) 708-720.
[6]
G. Loewenstein, J. S. Lerner, The role of affect in decision making, Handbook of affective science. (2003) 619-642.
[7]
J.R. Averill, Studies on anger and aggression: Implications for theories of emotion, American Psychologist. 38.11 (1983) 1145-1160.
[8]
L.L. Carstensen, B. Turan, S. Scheibe, N. Ram, H. Ersner-Hershfield, G. R. Samanez-Larkin, K. P. Brooks, J. R. Nesselroade. Emotional experience improves with age: evidence based on over 10 years of experience sampling, Psychology and aging. 26.1 (2011) 21-33.
[9]
D. McDuff, R. El Kaliouby, D. Demirdjian, R. Picard, Predicting online media effectiveness based on smile responses gathered over the internet, Automatic Face and Gesture Recognition (FG). 10th IEEE International Conference and Workshops. (2013) 1-7.
[10] R. D. Lane, P. ML. Chua, R. J. Dolan, Common effects of emotional valence, arousal and attention on neural activation during visual processing of pictures, Neuropsychologia. 37.9 (1999) 989-997. [11] R. J. Dolan, Emotion, cognition, and behaviour, Science. 298.5596 (2002) 1191-1194. [12] J.D. Mollon, "“Tho'she kneel'd in that place where they grew…” The uses and origins of primate color vision." Journal of Experimental Biology. 146.1 (1989) 21-38. [13] T. Hatta, H. Yoshida, A. Kawakami, M. Okamoto, Color of computer display frame in work performance, mood, and physiological response, Perceptual and motor skills. 94.1 (2002) 39-46. [14] R. Mehta, R. J. Zhu, Blue or red? Exploring the effect of color on cognitive task performances, Science. 323.5918 (2009) 1226-1229. [15] A.J. Elliot, M. A. Maier, Color and psychological functioning, Current Directions in Psychological Science. 16.5 (2007) 250-254. [16] K.W. Jacobs, F. E. Hustmyer Jr, Effects of four psychological primary colors on GSR, heart rate and respiration rate, Perceptual and motor skills. 38.3 (1974) 763-766. [17] N. P. Juslin, K. R. Scherer, Vocal expression of affect, The new handbook of methods in nonverbal behavior research. (2005) 65-135. [18] C. Caffi, R.W. Janney, Toward a pragmatics of emotive communication, Journal of Pragmatics. 22(3–4), (1994) 325–373. [19] N.J. Stone, A.J. English, Task type, posters, and workspace color on mood, satisfaction, and performance, Journal of Environmental Psychology. 18 (1998) 175–185. [20] R.A. Hill, R. A. Barton, Psychology: red enhances human performance in contests, Nature. 435.7040 (2005) 293-293. [21] T. Gnambs, M. Appel, B. Batinic, Color red in web-based knowledge testing, Computers in Human Behavior. 26.6 (2010) 1625-1631. [22] C.A. Thorstenson, Effects of Color Perception and Enacted Avoidance Behavior on Intellectual Task Performance in an Achievement Context, Diss. Appalachian State University, 2012. [23] R. Bagchi, A. Cheema, The effect of red background color on willingness-to-pay: The moderating role of selling mechanism, Journal of Consumer Research. 39.5 (2013) 947-960. [24] M. Bar, A cortical mechanism for triggering top-down facilitation in visual object recognition, Journal of Cognitive Neuroscience. 15.4 (2003) 600-609. [25] M. Bar, M.Neta, H. Linz, Very first impressions, Emotion. 6.2 (2006) 269-278. [26] M. Bar, M. Neta, Humans prefer curved visual objects, Psychological science. 17.8 (2006) 645-648. [27] L.A. Zebrowitz, Reading faces: Window to the soul?, New Directions in Social Psychology, Westview Press, Boulder, CO, 1997. [28] J. Aronoff, B. A. Woike, L. M. Hyman, Which are the stimuli in facial displays of anger and happiness? Configurational bases of emotion recognition, Journal of Personality and Social Psychology. 62.6 (1992) 1050-1066. [29] M. Bar, The proactive brain: using analogies and associations to generate predictions, Trends in cognitive sciences. 11.7 (2007) 280-289. [30] M.M. Bradley, P. J. Lang, Measuring emotion: the self-assessment manikin and the semantic differential, Journal of behavior therapy and experimental psychiatry. 25.1 (1994): 49-59.