Switching between different cognitive strategies induces switch costs as evidenced by switches between manual and mental object rotation

Abstract

Switching between tasks incurs costs when compared to repeating the same task. It is unclear whether switch costs also occur when repeating the same task but switching the under-lying cognitive mechanism (CM). Here, we investigated whether such CM switch costs exist and whether possible CM switch costs impact how frequently participants switch between dif-ferent CMs. Specifically, we asked participants to judge the identity of two misaligned objects by either mental or manual computer-mediated object rotation. While the specific CM (mental or manual) participants should use in a given trial was indicated by a cue in a first phase, it was freely chosen in a second phase. Participants exhibited considerable CM switch costs both for cued and freely chosen switches as indicated by increased reaction time. Moreover, individual switch costs moderately predicted switch frequency in the free choice condition, while an over-all tendency for CM perseveration was clearly observed. We conclude that CM perseveration can be beneficial from a processing time and likely also from a cognitive effort perspective, and that this benefit is among the causes for the often observed tendency to perseverate on one CM when several CMs are available.

Perseveration on Cognitive Strategies

“…However, when only comparing the present experiment with results from the same paradigm, differences vanish (i.e., 41% vs 43%, compare Figure 1g). However, in that study, participants showed the opposite perseveration pattern than in the present experiment: the iCM was preferred over the eCM (Figure 1g; Weis & Kunde, 2022). This opposite perseveration pattern might have been due to RT performance differences: While iCM use was slower than eCM use in the present study, iCM use was faster than eCM use in the former study 7 , which supports the idea that RT differences could be the driving force behind perseveration.…”

“…Such perseverations on one of both mechanisms are reflected by a bimodal distribution. In particular, data was obtained from joint arithmetic and social problem solving with a telepresent agent (a; Weis & Herbert, 2022), from solving alphanumeric problems with or without help from a computer (b; Weis & Wiese, 2019a), from intention offloading with or without spatial manipulation on a computer screen (c-f Scarampi & Gilbert, 2020), and from an object rotation task with mental or manual computer-mediated rotation (g; ` Weis & Kunde, 2022). Dashed gray lines represent means.…”

“…Why would problem solvers exhibit such partly extreme perseveration on a specific CM? There are a multitude of reasons, including the minimization of cognitive effort induced by switch costs (Dunn & Risko, 2019;Kool et al, 2010;Schillemans et al, 2012;Weis & Kunde, 2022), metacognitive biases (e.g., Gilbert et al, 2022;Risko & Dunn, 2015;Touron, 2015;Weis & Wiese, 2019c), and performance considerations based on speed, accuracy, or reliability of available CMs (e.g., Gilbert, 2015;Gray et al, 2006;Weis & Wiese, 2019c). With the present study, we aimed at investigating whether problem solvers are perserverating even in situations where iCM and eCM exhibit highly similar performances.…”

“…Second, and closely related, we assumed some participants to perseverate to avoid the choice costs associated with deciding which CM to use as well as the costs associated with working with two rather than one CM throughout a block. Such switch-related costs have been associated with cognitive effort and demand on executive functions (Kool et al, 2010) and were found both for voluntary task switches (Arrington & Logan, 2004;Kiesel et al, 2010) and voluntary CM switches (Weis & Kunde, 2022). The generally low level of perseveration after calibration makes choice cost avoidance an unlikely reason for perseveration.…”