Recognition of letters in words and nonwords.

Two visual search experiments were conducted to clarify the processes underlying the "word superiority" effect. Ss in both experiments (18 college students in Experiment 1; 18 college students and 18 older adults in Experiment 2) searched six, 50-item stimulus lists for the letters "a" and "r". Each item consisted of a string of four letters forming either a word or a nonword anagram of that word. Items were printed in either uppercase letters, lowercase letters, or mixed-cased letters (e.g., wOrD), resulting in the six stimulus lists. Significant main effects were obtained in the two experiments. Search times were shorter for words than for nonwords, regardless of case. Thus, the word superiority effect was demonstrated even when the visual configuration of the word was disrupted. Search times were also shorter for uppercase and lowercase items (i.e., same-case items) than for mixed-case items, regardless of the word/nonword condition. In addition, older persons (Experiment 2) evidenced slower processing than college students in all stimulus conditions, and their processing was differentially slowed in conditions which prevented automatic (as opposed to effortful) processing. Results were interpreted in terms of automatic information processing.