Long-standing questions in human perception concern the nature of the visual features that underlie letter recognition and the extent to which the visual processing of letters is affected by differences in alphabets and levels of viewer expertise. We examined these issues in a novel approach using a same–different judgment task on pairs of letters from the Arabic alphabet with 2 participant groups: 1 with no prior exposure to Arabic and 1 with reading proficiency. Hierarchical clustering and linear mixed-effects modeling of reaction times and accuracy provide evidence that both the specific characteristics of the alphabet and observers’ previous experience with it affect how letters are perceived and visually processed. The findings of this research further our understanding of the multiple factors that affect letter perception and support the view of a visual system that dynamically adjusts its weighting of visual features as expert readers come to more efficiently and effectively discriminate the letters of the specific alphabet they are viewing.
Small-N designs (SND) pose various challenges to statistical analysis. However, advances in statistical methods and computing power have led to a renewed interest in statistical analysis of SND. This paper focuses on a multiple regression approach, linear mixed-effects modeling (LMEM), arguing that this class of regression models addresses statistical pitfalls of ordinary least squares regression as applied to SND. Although LMEM analysis of longitudinal, repeated-measures designs is not new, it has not been extensively examined for SND intervention studies. Here we use LMEM to analyze both continuous (reaction time) and binomial (accuracy) measurement data from a training study of five individuals with dysgraphia subsequent to stroke, investigating the robustness of LMEM in terms of Type I error inflation, statistical power, and autocorrelation. The findings reveal that this approach to repeated-measures SND is robust with as few as five participants, for both continuous and binomial dependent measures. The results are presented with guidance on implementing an LMEM approach, with recommendations on how to achieve robust models.
Little is known about changes in neural representations following post-stroke recovery. Recent work suggests that the local heterogeneity of neural responses may index the degree to which representations are more or less broadly distributed within a region (Jiang et al., 2013), with greater heterogeneity indicating more highly-tuned, compact representations. We introduce a novel technique – Local-Heterogeneity Regression (Hreg) Analysis - to quantify local neural heterogeneity. We apply this approach to acquired dysgraphia focusing on left ventral occipitotemporal cortex (vOTC) that has been associated with learning orthographic representations, testing the following hypotheses: 1) Pre-training local-Hreg values in vOTC will correlate with spelling accuracy; 2) Recovery of spelling will be indexed within vOTC by increases in heterogeneity
A study with readers fluent in one or two scripts (Arabic and/or Roman) reveals that in second-script learning the visual system develops new weightings of visual features. Furthermore, processing of visual complexity changes with increasing expertise, such that letters with more features are processed more effectively than those with fewer.
Recent work has shown that certain visual features are more important relative to others for processing letter-shapes, and that one of the hallmarks of expertise is the adjusting of this relative importance (e.g. Fiset et al., 2008; 2009). Furthermore, the importance ranking of visual features has been found to differ between expert and naïve viewers of the Arabic alphabet (Wiley, Wilson, & Rapp, under review). There is an open question as to whether biscriptal individuals thus permanently alter their weighting of visual features, or else modulate it in an online manner according to which script is currently being viewed. In this study, biscriptal readers of Arabic and English completed same-different judgments on pairs of letters from both alphabets, as did monoscriptal readers English. The relative importance of a set of 14 visual features present in both scripts was determined to differ depending on both expertise and alphabet, indicating flexible weighting of the visual features. The results suggest that both specificities of the set of stimuli currently being viewed and accumulated expertise determine which features will be highly ranked. An additional finding on the role of expertise reveals that more complex letters (as measured by the total number of visual features) are processed with more effort than simpler letters for the naïve viewer, an effect which not reverses with increasing experience such that complexity is a predictor of significantly faster and more accurate responses by experts viewers.
Single case designs (SCD), studies with single participants or small N groups, pose various challenges to statistical analysis and, consequently, until recently, visual analysis has been the preferred approach for assessing treatment effects in such designs. However, advances in statistical methods have led to a renewed interest in alternatives to visual analysis of SCD, with a number of new proposals appearing in both education and neuropsychological research. This paper focuses on the approach of mixed-effects modeling with crossed random effects for both participants and items (Baayen et al., 2008), arguing that this class of regression models addresses the statistical pitfalls of ordinary least squares regression as applied to SCD, and provides an important alternative to visual analysis.
While considerable research has examined the optimal intensity of aphasia treatment (Robey, 1998), there has been little work on the optimal distribution of practice within the treatment period. Research on learning and memory indicates that studied material is remembered longer when the same amount of study is distributed across multiple sessions rather than being concentrated (Pashler, et al, 2007). In this study we examine the effectiveness of distributed compared to “clustered” treatment schedules for individuals with acquired dysgraphia. Using the multiple regression approach of generalized Linear Mixed-Effects Models (LMEMs) (Barr et al., 2013) we evaluate the effectiveness of training words according to different training schedules. In addition to modeling the main effects of treatment while controlling for variables like word length and frequency, LMEMs address problems in repeated measures designs such as uneven spacing of measurements (Barr et al., 2013) and can take into account random variability in the treatment items. Thus, LMEM has a number of features that make this a promising approach to evaluating data from rehabilitation studies.
We report evidence that the orientation/location of visual features mediating letter identification predicts reaction time and accuracy in same/different judgments on Arabic letters. Comparison of naïve and expert groups reveals that experience alters both post-perceptual similarity and the relative importance of various feature types.
A long-standing question in vision concerns the contribution to visual perception of the viewer’s expertise with the stimulus domain and the role of non-visual factors, such as motoric or phonological knowledge. We examined these issues with a same-different judgment task on pairs of characters from the Arabic alphabet, following the procedure of Courrieu, Farioli, and Grainger (2004). Two types of participants performed the task: ones who had no previous experience with Arabic, and ones who were native speakers or advanced students. The same analyses were conducted for both groups. Comparisons of the principal components and hierarchical clusters derived from each group reveal clear effects of learning and experience on the visual perception of the letter shapes. The results demonstrate how knowledge of letter name/identity and motoric stroke patterns modify the perceptual space for even relatively simply visual objects.
A retrospective on the challenges of teaching Arabic in a public high school, including in terms of curriculum development and difficulties encountered by the students in acquiring Arabic as a second language.