COMPREHENSION OF PICTORIAL SYMBOLS - EFFECTS OF CONTEXT AND TEST METHOD

This research examined two factors involved in the evaluation of picto rial symbol comprehension: context (absence vs. presence of photograph s depicting the probable environments where a symbol would be seen) an d test method (multiple-choice with less vs. more plausible distracter alternatives vs. open-ended). We tested 33 pictorial symbols from var ious sources. The results showed that the multiple-choice test with le ss plausible distracters inflated comprehension scores by an average o f 30% compared with the other two tests, which did not differ. The pre sence of context increased symbol comprehension in the open-ended test and in the multiple-choice test that had more plausible distracters. Extensive preliminary procedures demonstrated the difficulty of formin g a multiple-choice test with plausible distracter alternatives. This fact, combined with multiple-choice tests' low ecological validity in reflecting the real-world task of symbol comprehension, suggests that this test should be avoided in favor of an open-ended testing procedur e. It is suggested that context provides ecologically valid cues that limit the range of possible constructs that the pictorial symbol could be, raising comprehension scores. The use of context may help reduce the costs (money, time, effort) of producing pictorial symbols with ac ceptable, above-criterion comprehension levels.