Cognitive processes of letter and digit identification

letter identification;digit identification;word recognition;letter position coding;reading;Cognitive Science

Schubert, Teresa MarieEgeth, Howard E. ;

Johns Hopkins University

关键词: letter identification;    digit identification;    word recognition;    letter position coding;    reading;    Cognitive Science;   

Others  :  https://jscholarship.library.jhu.edu/bitstream/handle/1774.2/37937/SCHUBERT-DISSERTATION-2015.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: JOHNS HOPKINS DSpace Repository

【 摘 要 】

Letter identification, the process of determining letter identities and their positions from a visual stimulus, has a long research history.Debate remains concerning the levels of representation involved, as well as the manner in which letter position is coded: letters must be specified at particular positions.Digit identification, a cognitive process with arguably parallel goals to letter identification, has been less studied.A central question concerns overlap between letter and digit identification processes: Given the similarities between digits and letters as visual symbols, do they share a single identification process or does processing differ based on stimulus category?This thesis combines two investigations, concerning the letter position code employed in identification, and the extent of overlap of digit and letter processing, including variables determining letter and digit identification speed, accuracy, and error patterns.To address the first topic, I elicit letter perseveration errors in a character identification task, seeking evidence for the type of position code which can successfully predict both letter position specificity and flexibility in word recognition.Consistent with prior results from other paradigms, conclusions from this study suggest a key role for representing letters based on their distance from both the beginning and end of the word.Addressing the second topic, I explore the ability of character frequency, visual similarity, and visual complexity of letters and digits to predict a categorical distinction in identification performance.In behavioral experiments with unimpaired readers, I elicit identification errors for mixed letter and digit strings and same/different judgment response times to pairs of characters, providing a rich empirical dataset on which to further test the influence of these variables.The results have implications for our knowledge of letter and digit processing, and are consistent with a shared identification system in which both character types are identified.Together these investigations inform cognitive processes of letter and digit identification, which is the first stage in the reading process.Reading is a skill which is remarkably automatic, yet not innate, and the study of reading introduces broad questions about the nature of cognitive and neural processes that evolved to solve other problems in visual cognition.