Joseph Sabine, Teki Sundeep, Kumar Sukhbinder, Husain Masud, Griffiths Timothy D
Institute of Cognitive Neuroscience, University College London, UK; Institute of Neurology, University College London, UK.
Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford OX1 3QX, UK.
Brain Res. 2016 Jun 1;1640(Pt B):183-92. doi: 10.1016/j.brainres.2016.01.044. Epub 2016 Feb 2.
Auditory working memory (WM) is the cognitive faculty that allows us to actively hold and manipulate sounds in mind over short periods of time. We develop here a particular perspective on WM for non-verbal, auditory objects as well as for time based on the consideration of possible parallels to visual WM. In vision, there has been a vigorous debate on whether WM capacity is limited to a fixed number of items or whether it represents a limited resource that can be allocated flexibly across items. Resource allocation models predict that the precision with which an item is represented decreases as a function of total number of items maintained in WM because a limited resource is shared among stored objects. We consider here auditory work on sequentially presented objects of different pitch as well as time intervals from the perspective of dynamic resource allocation. We consider whether the working memory resource might be determined by perceptual features such as pitch or timbre, or bound objects comprising multiple features, and we speculate on brain substrates for these behavioural models. This article is part of a Special Issue entitled SI: Auditory working memory.
听觉工作记忆（WM）是一种认知能力，它使我们能够在短时间内积极地在脑海中保留并操控声音。基于对与视觉工作记忆可能存在的相似之处的考虑，我们在此针对非言语听觉对象以及时间，提出了一种关于工作记忆的独特观点。在视觉领域，关于工作记忆容量是局限于固定数量的项目，还是代表一种可在项目间灵活分配的有限资源，一直存在激烈的争论。资源分配模型预测，随着工作记忆中所维持项目总数的增加，项目表征的精度会降低，因为有限的资源在存储对象之间共享。我们在此从动态资源分配的角度，考虑关于不同音高的顺序呈现对象以及时间间隔的听觉研究。我们思考工作记忆资源是否可能由诸如音高或音色等感知特征决定，或者由包含多种特征的绑定对象决定，并推测这些行为模型的脑基质。本文是名为“SI：听觉工作记忆”的特刊的一部分。
