Effects of Contextual Color on Recall: Border Color as a Lesson and Posttest Cue for Factual and Conceptual Information Presented in Computer-based Instruction

Open Access
- Author:
- Prestera, Gustavo Eduardo
- Graduate Program:
- Instructional Systems
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- May 01, 2003
- Committee Members:
- Barbara L Grabowski, Committee Chair/Co-Chair
Roy Clariana, Committee Chair/Co-Chair
Francis M Dwyer Jr., Committee Member
Michael Mc Neese, Committee Member - Keywords:
- encoding specificity
screen design
automatic encoding
constructed response
fill-in items
item difficulty
gender differences
attention arousal
color affordances
outshining hypothesis
color effects
recall
user interface design - Abstract:
- This investigation examined the effects of contextual border colors on recall of the to-be-remembered information, when border colors were displayed on-screen during a computer-based instruction module. The encoding specificity principle (Tulving & Thomson, 1973) posits that contextual cues can only be useful during the posttest if they have first been encoded during the lesson. It was hypothesized that color (a context cue) would enhance recall when present in both the lesson and the posttest. Conversely, it was predicted that color would inhibit recall when the color hues in the posttest were intentionally mismatched with those in the lesson. It was also predicted that color effects would differ by gender, item type, and item difficulty. To examine the effects of border colors, a posttest-only control group design was used, with lesson condition and posttest condition as the crossed independent variables, with gender as an additional between-groups independent variable, and with two repeated measures: item type and item difficulty. The dependent variable was cued recall measured with fill-in-the-blank posttest items. An additional outside control group was also used to control for color motivational effects. Participants were 196 undergraduate students randomly assigned to five treatment groups. Participants read four passages, completed a posttest, read another four passages, and completed a second posttest. The order of reading passages was crossed to offset sequence effects. Depending on the treatment condition, participants saw color-coded border colors in both the lesson and the posttest, no border colors in either, border colors in the lesson and not the posttest, and vice-versa. A fifth group, an additional outside control group, saw border color hues in the posttest that were intentionally mismatched with those seen in the reading passages. The results do not support the hypothesis that color needs to be encoded during the lesson in order to enhance retrieval, contrary to encoding specificity. The presence of border colors in the lesson had positive, non-significant effects on recall of factual and conceptual knowledge, as measured by a difficult fill-in-the-blank posttest. However, the presence of border colors in the posttest varied significantly by item type, item difficulty, and gender. Recall of easy factual knowledge cued with verbatim posttest items was significantly higher when border colors were present in the posttest, whether or not border colors were present in the lesson. For males, the presence of border colors in the posttest significantly enhanced recall of the to-be-remembered information. For females, the presence of border colors in the posttest inhibited recall, though this negative effect did not consistently reach significance. These effects are discussed in relation to gender, color, HCI, and context-dependency research.