Unfolding visual lexical decision in time

Abstract

Visual lexical decision is a classical paradigm in psycholinguistics, and numerous studies have assessed the so-called "lexicality effect" (i.e., better performance with lexical than non-lexical stimuli). Far less is known about the dynamics of choice, because many studies measured overall reaction times, which are not informative about underlying processes. To unfold visual lexical decision in (over) time, we measured participants' hand movements toward one of two item alternatives by recording the streaming x,y coordinates of the computer mouse. Participants categorized four kinds of stimuli as "lexical" or "non-lexical:" high and low frequency words, pseudowords, and letter strings. Spatial attraction toward the opposite category was present for low frequency words and pseudowords. Increasing the ambiguity of the stimuli led to greater movement complexity and trajectory attraction to competitors, whereas no such effect was present for high frequency words and letter strings. Results fit well with dynamic models of perceptual decision-making, which describe the process as a competition between alternatives guided by the continuous accumulation of evidence. More broadly, our results point to a key role of statistical decision theory in studying linguistic processing in terms of dynamic and non-modular mechanisms.

Figure 1. Participants' performance on a visual lexical decision task, in which responses were modulated by the lexicality of the stimuli in accuracy parameters and overall lexical decision time.

Error bars depict Standard Error of the mean. a) Average time of movement initiation; b) Average lexical decision reaction time; c) Percentage of categorization errors; d) Percentage of responses after the deadline.

Figure 2. Real-time mouse trajectories.

A) Trajectories for High and Low Frequency words. Correct category is on the left and the opposite category is on the right. Trajectories for LF words showed attraction to the ‘nonlexical’ response alternative, which was statistically significant as indexed by AUC (bar plot). B) Trajectories for Pseudowords and Letter Strings. The correct category is on the right and the opposite category is on the left. Trajectories for Pseudowords showed attraction to the ‘lexical’ response alternative, which was statistically significant as indexed by AUC (bar plot).

Figure 3. Distributions of trajectory curvature.

a) Overlaid histograms of trajectory curvature for Lexical stimuli as measured by z scored values of AUC. High Frequency words exhibit unimodal distribution, whereas Low Frequency stimuli exhibit bimodality, with a first local maxima between a −.8 and −.4 z-score, and a second smaller mode between a .4 and .8 z-score. b) Overlaid histograms of trajectory curvature for Non Lexical stimuli as measured by z scored values of Area Under the Curve. Letters Strings have unimodal distribution. Pseudowords show bimodality with the first local maxima between a −.1 and −.4 z-score, and a second smaller mode between a .2 and .6 z-score.

Figure 4. Intercategory difference score.

Trajectory difference between Lexical and Nonlexical stimuli (dotted line). Averaged movement trajectories show difference between Pseudowords and Letter strings starting at the 43^rd normalized time (345 msec post-stimulus appearance), reaching the maximum amplitude at the 66^th time slice (690 msec post-stimulus appearance). Not much difference emerged for Lexical stimuli.

Figure 5. Line charts exhibit stimuli distribution along a “lexicality dimension".

a) Percentage of Item Accuracy (categorization errors and Out of Time trials), increasing stimulus ambiguity resulted in higher percentage errors; b) Trajectories Curvature Areas, more pronounced for Pseudowords, which deviate from the ideal straight response line; c) Trajectories Maximum Deviation, more pronounced for Pseudowords, which deviates from the ideal straight response line.