The case against specialized visual-spatial short-term memory.

Abstract

The dominant paradigm for understanding working memory, or the combination of the perceptual, attentional, and mnemonic processes needed for thinking, subdivides short-term memory (STM) according to whether memoranda are encoded in aural-verbal or visual formats. This traditional dissociation has been supported by examples of neuropsychological patients who seem to selectively lack STM for either aural-verbal, visual, or spatial memoranda, and by experimental research using dual-task methods. Though this evidence is the foundation of assumptions of modular STM systems, the case it makes for a specialized visual STM system is surprisingly weak. I identify the key evidence supporting a distinct verbal STM system-patients with apparent selective damage to verbal STM and the resilience of verbal short-term memories to general dual-task interference-and apply these benchmarks to neuropsychological and experimental investigations of visual-spatial STM. Contrary to the evidence on verbal STM, patients with apparent visual or spatial STM deficits tend to experience a wide range of additional deficits, making it difficult to conclude that a distinct short-term store was damaged. Consistently with this, a meta-analysis of dual-task visual-spatial STM research shows that robust dual-task costs are consistently observed regardless of the domain or sensory code of the secondary task. Together, this evidence suggests that positing a specialized visual STM system is not necessary. (PsycINFO Database Record

Cognitive Offloading: Structuring the Environment to Improve Children's Working Memory Task Performance

“…Clearly the task involved visuospatial information, so we might have expected a stronger relationship here. However, some working memory models eschew the idea of domain specific storage, instead proposing a single "focus of attention" (e.g., Cowan, 1999Cowan, , 2005, or at least argue against the need to include a specialized visual storage capacity (Gray et al, 2017;Morey, 2018; but see Baddeley, Hitch, & Allen, 2019;Hanley & Young, 2019). Indeed, even models that contain modality-specific sub-components acknowledge that working memory is an inter-linked, holistic construct (Baddeley, 2012;Logie, 2011).…”

Flexible representations in visual working memory and interactions with long‐term learning: Commentary on the special issue

“…The past 20–30 years have seen a surge of interest in visual working memory (VWM), and it has become a lively field with many, often overlapping, topics of investigation (and debate) concerning: whether VWM is limited by the number of items that can be held (Adam, Vogel, & Awh, ; Cowan, ; Luck & Vogel, ) or by the distribution of a flexible resource (Bays, Catalao, & Husain, ; van den Berg, Shin, Chou, George, & Ma, ); whether representation in VWM is object‐based and how features are retained bound in VWM (see Schneegans & Bays, this issue, for a review); how structure and redundancy in to‐be‐remembered material influences performance (Brady & Alvarez, ; Liesefeld, Liesefeld, & Müller, this issue; Morey, this issue); to what extent storage of visual material occurs separately from storage of other material (Berggren & Eimer, this issue; Logie, ; Morey, ); what role VWM plays in real‐world looking behaviour and visual search (Annac, Zang, Müller, & Geyer, this issue; Berggren & Eimer, this issue; Pollmann, this issue); and questions about the neurobiology underlying VWM (for a review of oscillatory underpinnings, see Sauseng, Peylo, Biel, Friedrich, & Romberg‐Taylor, this issue) and what factors influence its capacity, such as the emotional valence of items (Curby, Smith, Moerel, & Dyson, this issue) or exercise (Dodwell, Müller, & Töllner, this issue).…”

Perceptual grouping boosts visual working memory capacity and reduces effort during retention

“…The effects of feature redundancies and similarities seem to differ for visual and auditory‐verbal materials. Because short‐term memory phenomena for serial verbal materials have been studied so extensively, a natural default assumption for visual materials is to assume that the same phenomena would appear in visual memory (Morey, ). This leads to the expectation, for instance, that visually similar items will be confused with each other and their presence in a stimulus will lead to increased errors, as with acoustically similar stimuli (Baddeley, ; Conrad, ).…”