CHAPTER 458 55 54 53 52 51 50 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 30 Working Memory, Deafness, and Sign Language Matt Hall and Daphne Bavelier A review of the studies that compare short-term memory (STM) in deaf and hearing populations highlights an advantage for speech in STM tasks using digit or letter span, but not in those using visuospatial stimuli. he indings have remained relatively consistent over time, but their interpreta- tion has been less clear. Some have argued that those working memory (WM) components that process visuospatial information also process sign language, whereas others have suggested that the diferences can be explained by factors like articulatory dura- tion or simple auditory deprivation (for a discussion see Wilson, Bettger, Niculae, & Klima, 1997). Still others have proposed that signers are impaired when it comes to retaining information about temporal order but advantaged for remembering spatial loca- tions (for a discussion see Bavelier, Newport, Hall, Supalla, & Boutla, 2008). We posit that the data are best captured by recognizing that signers and speakers share the same WM architecture, but preferentially rely on diferent WM subprocesses. Speakers depend on the phonological loop to a much greater extent than signers, who in turn rely on distributed coding across phonological, visuospatial, and also episodic processes. According to this view, the speech-sign diferences in memory tasks are not indicative of fundamental diferences in WM capacity between deaf and hearing people, but rather relect diferent processing biases in WM as a function of auditory deprivation and language exposure. We therefore urge educators and clinicians to be cautious when interpreting the results of standardized tests, and call on researchers to give due attention to the less- investigated aspects of WM in both signers and speakers. Abstract Working memory (WM) refers to the human capacity to encode, store, manipulate, and recall information. A proper understanding of WM therefore provides essential insights into human cognition. This chapter reviews available research concerning the impact of deafness and sign language use on WM, much of which comes from the study of a single subcomponent of WM termed short-term memory (STM). We argue that excessive focus on STM (the ability to encode, store, and retrieve a sequence of unrelated words in serial order) to the exclusion of other WM subprocesses has caused an extreme interest in phonological coding at the expense of other known codes used in WM, in particular visual but also episodic codes. Deafness and use of a sign language may result in greater reliance on not only visual but also episodic coding, as compared to what is typically observed in hearing nonsigners. This multiple coding hypothesis calls into question whether the robust phonological bias described in hearing individuals should be taken by researchers, clinicians, and educators as the gold standard for deaf populations. Keywords: working memory , deafness, sign language, short-term memory , digit span, phonological store, multiple coding hypothesis, episodic buffer 30-Marschark-Chap_30.indd 458 30-Marschark-Chap_30.indd 458 2/8/2010 3:33:01 PM 2/8/2010 3:33:01 PM OUP UNCORRECTED PROOF – FIRST-PROOF, 08/02/2010, GLYPH