Eye movements as indicator of solution knowledge in anagram solving

The current study attempted to replicate the results of Ellis et al. (2011) in an extended way. In the original study Ellis et al. demonstrated that eye movements may indicate the solution knowledge prior to response. They compared viewing times on the distractor letter vs. solution letters. Viewing time on distractor letter started decreasing since several seconds prior to response both in insight and non-insight trials. We added two additional parameters of anagrams: solution-word frequency and frequency of cooccurrence (“agglutination”) of the distractor letter with the solution letters. Low-frequency words and/or stimuli with high agglutinating distractor were solved less often and longer than others, with the effect of distractor type only for the low-frequency words. Eye-tracking data analysis revealed that either in insight and non-insight trials distractor did not differ from solution letters in the first half of the solving process, but had fewer viewing time in the second half of the solving process. In the more difficult stimuli (by solution- word frequency and distractor type) distractor was revealed later than in easier ones or did not differ at all. Eye-tracking data on viewing time on distractor vs. solution letters were in accordance with the Ellis et al.’s results and the anagram difficulty factors.

Phasic Affective Modulation of Creativity

Thus far, it has been shown that positive compared to negative mood increases creativity, for instance, as measured by performance in the remote associate task, where participants are asked to find the common remote associate for three clue words (e.g., RABBIT CLOUD MILK solution word WHITE). The present experiments show that very brief variations in affect, lasting for only a few seconds and changing from trial to trial within participants, are sufficient to modulate creativity in that task, presumably by modulating the breadth of semantic spread. Using word valence of remote associates themselves (Experiment 1), and consonant and dissonant chords (Experiment 2) as affect inductions, it was observed that brief positive compared to negative affect increased creative performance. This evidence extends the affect-modulation hypothesis to a temporal micro-level.

Solving Anagrams: Category Priming and the Differential Availability of Category Solutions

An experiment was performed to test the hypothesis that the effect of category name priming on anagram solving varies with the strength of the relationship between the solution word and the priming category. Subjects solved anagrams of taxonomic category instances under primed or unprimed conditions. In the primed condition, the name of the taxonomic category from which the solution word was chosen was provided on each trial. Priming was shown to facilitate anagram solution and the extent of this facilitation was directly related to the instance dominance of the solution word in the priming category. The results were discussed in terms of current models of semantic memory.

Effects of Solution Word Attributes on Anagram Difficulty: A Regression Analysis

A sample of 45 student subjects provided solution scores for 80 five-letter anagrams. These scores were analysed as a function of solution word imagery, con-creteness, familiarity, objective frequency, age-of-acquisition and associative meaningfulness using multiple regression techniques. Two bigram measures together with number of vowels, nature of starting letter (vowel or consonant), anagram pronounceability and anagram-solution similarity scores were also entered into the regression equations. The bigram measures, the starting letter and anagram-solution similarity emerged as having significant associations with the solution scores. Previous reports of imagery effects in anagram are discussed in the light of the present results.

Performance of Male Schizophrenics on An Anagram-Solution Task: An Application of An S-R Mediation Model

2 groups of 15 volunteer schizophrenic Ss each worked on 20 word-anagram problems. The anagrams for Group I were the solutions for Group II, and vice versa. Each of the last 10 problem pairs represented a test of the mediation model's prediction that faster solution times will occur when the anagram has a low Thorndike-Lorge frequency and the solution word has a high Thorndike-Lorge frequency. Results strongly confirm this prediction. When anagram and solution were of equal frequency, faster solution times tended to be equally distributed among the two groups. Distinct deficits were evident in the performance of schizophrenics. Over-all results indicate the model is applicable for a schizophrenic sample.