Date of Award
Doctor of Philosophy
Erin F. MacDonald
Product visual designs convey a variety of information about these products to consumers. These designs play an important role in affecting consumer judgments, which further determine purchase decisions. Understanding how consumers decode visual designs to form judgments as well as how to use visual designs to affect consumer judgments are important. Insights in these will help designers make better design decisions and also present new possibilities of product design.
This dissertation employs eye-tracking technology to assist in understanding consumers' decoding processes. First, eye-tracking is used to examine how consumers evaluate visual designs to determine preferences and product differences. Then, eye-tracking is utilized to help investigate influences of (1) pairing products that have both commonalities and differences and (2) visual cues on consumer judgments separately.
Product features, defined as visible product attributes, are important constitutions of product visuals. Study 1 uses eye-tracking to address two topics about product features: (1) feature importance in preference decisions and (2) whether or not consumers can detect a feature's size change. Results from eye-tracking how subjects evaluated product images to determine preferences showed a feature's gaze data (e.g., how long the subjects fixated on the feature) significantly correlated with the feature's importance rating provided by the subjects. Results from eye-tracking how subjects detected differences between product images showed noticeable and unnoticeable feature size changes had significantly different corresponding gaze data. Statistical models of gaze data can predict importance and size change saliency of a feature.
Purchase decisions often require comparing products that have both commonalities and differences. Study 2 investigates how this configuration of choice alternatives influences consumer judgments by testing a model of choice from psychology, the cancellation-and-focus (C&F) model, in the product design domain. The C&F model specifies when facing two choice alternatives that have both shared and unique attributes, people tend to ignore the shared attributes and focus on the unique ones, which can affect both preferences and certain postpreference judgments. The model had only been tested with text-only stimuli, where text-described attributes represented products. Study 2 tested the model with image-only, text-only, and image-with-text stimuli separately. It tested each stimuli type with two conditions: (1) presenting stimuli sequentially and (2) side-by-side. The C&F model held only in limited situations for the tested products. Generally, the unique attribute/feature had more gaze attention than the shared one, indicating the importance of product differences in consumer preferences. While a shared attribute was canceled in decisions, a shared feature reinforced impressions.
Consumers extract cues from visual designs and mentally associate them with unobservable product attributes to aid judgments. Study 3 investigates the possibility to rapidly build mental associations to influence consumer judgments. The study also compares the effectiveness of cuing holistically, through body shapes, and cuing by features. Subjects participated in an association-building task, where a visual cue was associated with either a positive or a negative judgment of environmental friendliness. Results from a latter testing task demonstrated that mental associations between body shape cues and environmental friendliness formed. Body shape cues affected products' environmental friendliness ratings in the desired direction, but feature cues did not. Gaze data showed the subjects adjusted their distributions of attention to a product after the association-building task, indicating the ability of cues to promote a more efficient decision-making behavior.
Du, Ping, "Investigating effects of product visual designs on consumer judgments with the aid of eye-tracking" (2016). Graduate Theses and Dissertations. 15122.
Available for download on Thursday, April 20, 2017