Eye fixation patterns in the solution of mathematical word problems by young adults: Relation to cognitive style and spatial abilit

Eye tracking technology was used to investigate the relationship between problem solving behavior and three subject variables--spatial visualization ability (SVA), field dependence/independence, and mathematical achievement. Eye movement variables examined were average eye fixation duration and proportion of total fixation time spent on essential and nonessential information within six two-step mathematical word problems.;Eleven females and 26 males (median age: 20.3) enrolled in the second of two remedial mathematics courses at The Ohio State University (Math 075) comprised the final sample. Each subject completed Vandenberg's Mental Rotations Test (MRT) and the Group Embedded Figures Test (GEFT). Eye movements were monitored while subjects solved the six word problems.;Major results. (1) Average fixation durations on both extraneous and essential numerical information were significantly longer than average fixation durations on the question or on nonessential verbal information. (2) In the only problem where extraneous numerical information preceded the essential numerical information, field independence (FI) and SVA were negatively related to: (a) the proportion of time allocated to fixating the extraneous numerical information, and (b) the ratio of the percent of total fixation time on the extraneous numerical information to the percent of total fixation time on the essential information. In the same problem, positive relationships were found between: (a) SVA and the proportion of total fixation time on the essential numerical and verbal information, and (b) FI and the proportion of time spent on the question. (3) In the two problems where the extraneous numerical information was preceded by relevant numerical information, SVA was positively related to the average fixation duration on the essential numerical and verbal information. (4) FI was negatively related to the average fixation duration on the essential numerical and verbal information in each of the three problems containing extraneous information as well as in one of the three problems which did not. (5) The correlation between the GEFT and the MRT was.44.;SVA was positively related to a pattern of relatively few, long eye fixations on the essential information. In contrast, FI was positively related to a pattern of more numerous, but shorter eye fixations on the essential information. Possible explanations for these findings were discussed and implications for future studies were presented.