Paper features: A neglected source of information for letter recognition

Abstract

Adult observers have surprisingly low calculation efficiencies for letter recognition (see, e.g., Pelli, Burns, Farell, & Moore-Page, 2006). Here, we examine the possibility that this is partly due to observers' neglecting paper features (e.g., the absence of ascenders and descenders in 'o'). Each of 16 observers completed 5,000 trials of a single-letter two-alternative forced-choice detection task. Using a combination of classification image analyses and Bayesian statistical analyses, we argue that between 60% and 75% of our participants indeed neglected paper features.

Computational reconstruction of mental representations using human behavior

“…Averaging the stimuli associated with these "superstitious" detections reveals the mental representation of that signal, unbiased by external input. This approach has been used to recover representations of letters, facial expressions, and 3D patterns [8][9][10][11] . Findings from these and related methods informed and validated models and theories of letter identification, universality of facial expressions, and trustworthiness judgments, among other aspects of cognition [10][11][12][13] .…”

“…This approach has been used to recover representations of letters, facial expressions, and 3D patterns [8][9][10][11] . Findings from these and related methods informed and validated models and theories of letter identification, universality of facial expressions, and trustworthiness judgments, among other aspects of cognition [10][11][12][13] . One prominent disadvantage of this method however is that it typically requires thousands of trials per observer and is limited to artificial or simple target signals that always have the same pixel representation.…”

“…An additional drawback of reverse correlation studies is that they focus on reconstructing a limited number of predefined stimuli [8][9][10][11] . A more general, ambitious, and potentially fruitful approach would be to recover a function mapping the labels of all natural visual categories to visual features (see also refs.…”

Computational reconstruction of mental representations using human behavior

“…Averaging the stimuli associated with these "superstitious" detections reveals the internal representation of that signal, unbiased by external input. This approach has been used to recover representations of letters, facial expressions, and 3D patterns (Gosselin, Bacon & Mamassian, 2004;Gosselin & Schyns, 2003;Jack et al, 2012;Morin-Duchesne et al, 2014). However, it typically requires thousands of trials per observer and is limited to artificial or simple target signals that always have the same pixel representation.…”