Arousal, Memory, and Impression-Formation Effects of Animation Speed in Web Advertising
 
Student researchers

Sriram Kalyanaraman (PhD Student)
Christine Martin (BA Student)
Carson B Wagner (PhD Student, University of Colorado)

Faculty Supervisor

Dr. S. Shyam Sundar

For a complete report of this research, see:

Sundar, S. S., Kalyanaraman, S., Martin, C., & Wagner, C. B. (2001, May). Arousal, memory, and impression-formation effects of animation speed in Web advertising. Paper presented to the Information Systems Division at the 51st annual conference of the International Communication Association (ICA), Washington DC.

Introduction

Animated advertisements on the Web come in a variety of shapes, sizes and colors. But, they also animate at different speeds. While recent studies have shown animated ads to be more effective than still ads, the role played by the rate of motion in animated ads has been neglected. Given that the primary function of animation is to attract attention, such differences in speed of animation are likely to lead to differential levels of attention to the ads. Researchers have pointed out that speed is an obvious component of an interactive media system, and the degree to which it is realized in an interaction is bound to determine one's psychological experience with the system. An experiment was designed to address this issue by focusing specifically on the physiological and psychological effects of animation speeds in Web ads.

Hypotheses:

Several hypotheses, based on motion-effects, limited-capacity, and vividness-effects theories, were proposed.

H1: Fast animation ads will elicit greater physiological arousal than slow animation ads.

H2a: A slow animation ad will elicit greater physiological arousal when it follows, rather than precedes, a fast animation ad.

H2b: A fast animation ad will elicit greater physiological arousal when it follows, rather than precedes, a slow animation ad.

H3: Ad recognition for fast animation ads will be greater than ad recognition for slow animation ads.

H4: Ad recall for fast animation ads will be lesser than ad recall for slow animation ads.

H5a: A fast animation ad will elicit significantly greater positive impressions of the website when it follows, rather than precedes, a slow animation ad.

H5b: A slow animation ad will elicit significantly greater positive impressions of the website when it precedes, rather than follows, a fast animation ad.

Method

Forty-seven participants in a 2 (Animation) X 2 (Sequence) mixed-factorial experiment were exposed to two Web pages, each with a news story surrounded by four online advertisements. The speed (fast, slow) with which the ads were animated served as the within-participants factor, with the fast animation ads averaging 55 animations/flashes per minute and the slow animation ads averaging 21.5 animations/flashes per minute. The sequence (fast, slow; slow, fast) in which participants were exposed to fast and slow animation served as the between-participants factor. The primary dependent variable, physiological arousal, was measured by on-line recordings of skin conductance level (SCL). In addition, after exposure, participants filled out a paper-and-pencil questionnaire eliciting their behavioral intention toward the ads, their evaluation of the Web page as a whole, and their memory for details in the news stories as well as the ads on the pages.

Results

A series of 2x2 mixed ANOVAs was conducted with the two-category animation speed (fast, slow) variable as the within-participants factor and the two-category animation sequence variable (fast-slow, slow-fast) as the between-participants factor.

H1: Supported. When the ANOVA was performed with arousal (as indicated by percentage change in tonic skin conductance from baseline) as the dependent variable, a significant main effect for speed emerged, with study participants showing higher arousal levels while exposed to fast-animation ads than while responding to slow-animation versions of the same ads.

H2a: Not supported. The page with slow animation ads elicited greater arousal when it followed, rather than preceded, the page with fast animation ads, but a post-hoc test failed to show a statistically significant differentiation between these two means.

H2b: Supported. The Web page with fast animation ads elicited significantly greater arousal when it followed, rather than preceded, the page with slow animation ads.

H3: Not supported. There were no significant differences between the two speeds on the recognition measure.

H4: Not supported. There were no significant differences between the two speeds on the recall measure.

H5a: Not supported. The sequence in which participants were exposed to the page with fast animation ads did not affect their impressions of the page.

H5b: Disconfirmed. Participants found the Web page with slow animation ads more appealing when it was preceded by a page with fast animation ads than when it was not.

Conclusions

In summary, results from this experiment offer support to the notion that animation speed of Web ads is a psychologically relevant construct. People's responses to animation speed, however, seem to be overwhelmingly moderated by the sequence in which animation speeds are presented to them, as evidenced by the presence of significant interaction between the speed variable and the sequence variable on all analyses producing statistically significant results. Specifically, results confirm the hypothesized ability of fast animation ads to elicit significantly higher arousal then their slow counterparts, particularly when the fast ads are preceded by slow ones. However, on cognitive indicators and impression formation measures, psychological responses to fast animation ads are relatively unaffected by the sequence of presentation of ads with differential speeds whereas responses to slow animation ads, in general, are more positive when they are preceded by fast ads than when they are not.

For more details regarding the study contact

Dr. S. Shyam Sundar by e-mail at sss12@psu.edu or by telephone at (814) 865-2173

>> Return to abstract listings page
>> Return to the main research page

Media Effects Research Lab at College of Communications, Penn State University