![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
ABSTRACT
Two prominent cognitive capacity limitations are the maximal number of objects we can place in working memory (WM) and the maximal number of objects we can track in a display. Both are believed to have a numeric value of 3 or 4, which has led to the proposal that we have a general cognitive capacity, and that this capacity is most likely linked to limitations of how many objects we can attend simultaneously. Based on previous results showing that we can memorize more objects if they come from different categories than if they come from the same category (e.g., Feigenson & Halberda, 2008; Wong, Peterson, & Thompson, 2008; Wood, 2008), we compare how category-based grouping affects performance for WM and multiple object tracking (MOT). We present participants with either “pure” displays of either cars or faces, or with “mixed” displays of cars and faces. Overall, the effects of category are weak. In some analyses but not others, we replicate the mixed advantage for WM, albeit with a small effect size. In contrast, we observe a weak pure advantage for MOT tasks, at least in a meta-analysis of five experiments, but not in all experiments. Accordingly, WM and MOT tasks differed significantly in their sensitivity to category membership. We also find that WM is slightly better for faces than for cars, but that no such difference exists for MOT. We tentatively suggest that cognitive capacity limitations in different domains are at least partially due to the limitations of distinct mechanisms.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Trick & Pylyshyn (Citation1994) argued against a common mechanism between WM and subitizing, and argued that the spatial pointers called Fingers of Instantiations that allow us to track objects are pre-attentive. However, as we will discuss below, other authors argued that small-number processing might be related to WM via attention, and used MOT as a measure of attention.
2 In Fan & Turk-Browne's (Citation2016) experiments, participants first encoded either the position or the colour of a novel shape, a manipulation that affects LTM for the stimulus dimension that was encoded and probed (Fan & Turk-Browne, Citation2013). Following this encoding phase, participants completed a 1-item delayed match-to-sample task with monochrome versions of these shapes. Critically, in the retention interval, participants viewed two coloured Gabor patches and had to decide which of them was tilted. One of the patches had the colour that the memory object had during the encoding phase and the other patch a different colour. Participants were faster to detect the target (i.e., the tilted shape) if it shared the colour that the memory object had during the encoding phase, suggesting that LTM representations can guide attention.
3 In principle, it is possible that the very presence of distinct objects has a tracking cost compared to a homogenous tracking condition, because identifying the objects might distract resources from tracking even if the object identity is irrelevant for the task. This is particularly plausible because, at least in the case of WM, participants seem to encode the locations of objects relatively automatically (Makovski & Jiang, Citation2008). Empirically, however, tracking is improved if items have unique features (Makovski & Jiang, Citation2009a).
4 The significance level was somewhat smaller with a signed rank test, V = 2958, .
5 Given that, in some experiments, we observed weak mixed advantages for set-sizes 2 and 4, we also performed a meta-analysis for these set-sizes in the experiments where these set-sizes were administered (i.e., Experiments 1 and 3b, 44 participants in total). For set-size 2, the meta-analysis did not detect a deviation from chance,
,
,
,
. Entering the combined difference scores into a t-test did not detect a deviation from chance either (d = −.000,
),
,
,
,
; a signed rank test was not significant either, V = 75,
. Correcting with the BIC, the likelihood ratio in favour of the null hypothesis was 1.72. However, after correction with the AIC, we observed a likelihood ratio of 1.29 in favour of the alternative hypothesis (.78 in favour of the null hypothesis). We thus tentatively conclude that the significant mixed advantages we observed for set-size 2 were type I errors.
For set-size 4, the mixed-factor meta-analysis did not detect a deviation from chance, ,
,
,
,
. However, a combined t-test reached significance, (d = −.012,
),
,
, Cohen's
,
); a signed rank test was significant as well, V = 121,
. Further, likelihood analysis favoured the alternative hypothesis (
after BIC correction and
after AIC correction). It is thus unclear whether there is a mixed advantage at set-size 4.
6 A signed rank test did not differ from chance either, ,
.
7 A signed rank test differed from chance as well, V = 1378, .
8 A signed rank test was significant as well; V = 1011, .
9 While performance did not differ between car and face trials in MOT experiments, we observed a small face advantage for WM experiments. However, the face advantage did not differ significantly between MOT and WM experiments, either in a t-test, ,
,
, or in a signed rank test,
, or when comparing counts of face vs. car vs. no advantages,
.
10 We draw this conclusion for the change detection paradigm where all items are presented simultaneously, as this paradigm provided the most consistent evidence for fixed visual WM capacity limitations. However, it is possible that the link between MOT and WM might be stronger when the WM paradigm has a temporal component, for example when items are presented and tested sequentially (e.g., Endress & Potter, Citation2012; Intraub, Citation1980; Potter, Staub, Rado, & O'Connor, Citation2002).
11 For example, in Luck & Vogel's (Citation1997) change detection experiment, participants had to remember colour patches, such that 7 colours were reused across hundreds of trials. Hence, the participants faced the question to decide whether they had seen a given colour in the current trial, or one of the preceding trials, leading to proactive interference.
