|
Wisniewski, E.J. & Bassok, M. (1999). What makes a man similar to a tie? Stimulus compatibility with comparison and integration. Cognitive Psychology, 39, 208-238. Primary Reviewer: Angie Secondary Reviewer: Alex Purpose It is intuitive to assume that different types of processing are compatible with different types of stimuli. Specifically, stimuli such as apples and oranges, which are taxonomically related (i.e., they belong to the same superordinate category) are compatible with comparison, while stimuli such as apples and baskets, which are thematically related (i.e., apples go in baskets), are more compatible with integration. Stimuli that are taxonomically related are alignable (e.g., apple-orange). Alignability refers to the ease of comparison, the judgment of similarities and differences. Nonalignable stimuli (e.g., apple-basket, man-tie) may belong to categories that are based on thematic relations (e.g., events, places, scenes, etc.). The experiments in this article investigate the effects of stimulus compatibility on processing. The stimuli in these experiments are compatible on different levels; some are compatible only via comparison, others only via integration, some via both, or neither. By having subjects rate alignability and thematic relatedness (in separate experiments), the authors sought to examine the effects of such stimulus compatibility and the possible interaction between comparison and integration. The authors propose that the intuitive part of how different types of stimuli are processed differently may become counterintuitive regarding task-appropriate instructions. They believe single-process models overestimate the extent to which people’s performance is mediated by controlled, task-appropriate processing and not uncontrolled, stimulus-appropriate processing. The authors hypothesized that the stimulus compatibility of object pairs may override task-appropriate processing. Experimental Work Experiment 1 Participants were asked to rate the similarity of sets of object pairs on a scale of 1 to 7 (1 = very dissimilar, 7 = very similar). A subset of the participants were also asked to explain their ratings. It was predicted that when subjects made similarity judgments, different modes of processing (comparison vs. integration) would be employed depending upon the compatibility of the stimuli to the processes. In other words, pairs of objects that were taxonomically related would be judged as similar based on comparison. On the other hand, the similarity of pairs that were not taxonomically related would be judged based on integration. The explanations were examined to determine the processes used by participants when making the similarity judgments. Stimuli varied on two measures, leaving four possible conditions. The object pairs were either taxonomically related, and therefore highly alignable (e.g., milk-lemonade) or poorly alignable (e.g., milk-horse). The object pairs were also either thematically related (e.g., milk-cow) or not (e.g., milk-horse). The resulting conditions were: highly alignable and thematically related (e.g., milk-coffee); highly alignable and not thematically related (e.g., milk-lemonade); poorly alignable and thematically related (e.g., milk-cow); or poorly alignable and not thematically related (e.g., milk-horse). As predicted, similarity ratings for pairs that were highly alignable were reliably higher than for pairs that were poorly alignable (e.g., milk-lemonade vs. milk-horse), regardless of thematic relatedness. Similarity ratings for pairs that were thematically related were higher than for pairs that were not thematically related (e.g., milk-cow vs. milk-horse). Also, there was a significant interaction between alignability and thematic relatedness indicating that that the effect of thematic relatedness on similarity judgments was larger when the pairs were poorly aligned than when they were highly aligned. The similarity ratings given by participants who did not have to justify their ratings were nearly identical to the similarity ratings given by the subjects who did have to give explanations. This demonstrated that requiring justification did not change how subjects reacted to the object pairs and that even without needing to justify their responses, participants used both comparison and integration processes. Experiment 2 The purpose of Experiment 2 was to more closely examine the compatibility effects found in Experiment One. Participants were given explicit instructions to list commonalities and differences between the object pairs, eliciting use of the comparison process. Clear examples were given. The results were similar to Experiment 1 in that alignability and thematic relatedness strongly affected participants’ tendency to compare or integrate the object pairs. Experiment 3 Experiment 3 also examined the compatibility effects previously found. The task, though, was different in that participants were required to rate the extent to which the object pairs were thematically related. Again, clear examples were given. The results of this experiment mirrored the results of Experiments 1 and 2. Conclusions Several conclusions were drawn regarding this set of experiments. It was demonstrated that stimulus compatibility affects the outcomes of tasks that only involve one type of processing. Experiments 1 and 2 are typically believed to only involve comparison (similarity ratings, Experiment 1; commonalities and differences, Experiment 2). Experiment 3 should enlist only integrative processing (judging thematic relatedness). Clearly, though, the authors have shown that both integration and comparison are used in each of the tasks, regardless of what the exact task-appropriate behavior may be. There are many implications concerning the findings of the current research. Most processing accounts assume that each cognitive task is only mediated by one process and that the single process operates over the entire range of the stimuli. The above results clearly indicate otherwise. Psychological models of category formation focus only on comparison. Recent work, though, suggests that category formation is affected by both comparison and integration. Models of induction also focus solely on feature comparison, yet there is clear evidence that thematic relations play a role. There are also implications regarding metaphors, similes, and analogies and conceptual combination. The “take home point” here is that the current results indicate the need to extend cognitive models to include both integration and comparison processes and the effects of stimulus compatibility. Even when clearly instructed otherwise, people enlist both integration and comparison processes depending upon the stimuli. Points for Discussion A previously discussed question merits discussion again in terms of the current set of experiments. This question concerns participant compliance. Is it not possible that the participants in these experiments relied on integration when making similarity judgments because they believed that they needed to find some sort of similarity? Although explicit instructions were given, it does not seem out of the realm of possibility that subjects had an overriding “need” to indicate similarity on some sort of level. A related question arises regarding participant knowledge. Do they really understand the difference between thematic relatedness and comparison? Maybe they understood the directions, but believed that saying something in similar because of commonalities and differences means about the same thing as “similarity” due to thematic relatedness. In Experiment 1 (and elsewhere) an author and naïve coder divided justifications into instances of comparison and integration. There was 88% agreement between these two people in the first experiment. The disagreed upon 12% were resolved through discussion. Why were the questionable responses not just disregarded? If it is assumed that the naïve coder did not just mistakenly mislabel some of the justifications, was resolution through discussion really fair? Would the results have not have been significant? Also, this brings up just how muddied the line between integration and comparison may be at times.
|
|
This page was last updated: |