Control over sampling boosts numerical evidence processing in human decisions from experience

Abstract

When acquiring information about choice alternatives, decision makers may have varying levels of control over which and how much information they sample before making a choice. How does control over sampling affect the quality of experience-based decisions? Here, combining variants of a numerical sampling task with neural recordings, we show that control over when to stop sampling can enhance (i) behavioral choice accuracy, (ii) the build-up of parietal decision signals, and (iii) the encoding of numerical sample information in multivariate electroencephalogram (EEG) patterns. None of these effects were observed when participants could only control which alternatives to sample, but not when to stop sampling. Furthermore, levels of control had no effect on early sensory signals or on the extent to which sample information leaked from memory. The results indicate that freedom to stop sampling can amplify decisional evidence processing from the outset of information acquisition and lead to more accurate choices.

Over- and Underweighting of Extreme Values in Decisions from Sequential Samples

“…Lastly, the leakage parameter l was significantly larger than 0 (indicating greater weighting of later samples) in each of the experimental conditions (single-stream conditions: Z = 10.06, p < 0.001, r = 0.03, Wilcoxon signed-rank test against 0; dual-stream conditions: Z = 11.06, p < 0.001, r = 0.03, Wilcoxon signed-rank test against 0). Thus, we generally observed recency effects in our tasks, consistent with previous findings (Anderson, 1964;Appelhoff et al, 2021;Cheadle et al, 2014;Hubert-Wallander & Boynton, 2015;Spitzer et al, 2017;Summerfield & Tsetsos, 2015;Weiss & Anderson, 1969;Yashiro et al, 2020).…”

EEG-representational geometries and psychometric distortions in approximate numerical judgment

“…3b right ). Specifically, like in our previous work (Appelhoff et al, 2022; Spitzer et al, 2017), we observed neurometric estimates of k > 1 (i.e., anti-compression) and of b > 0 (i.e., a bias towards larger numbers; p=0.013, FDR-corrected), which mirrors the pattern observed in the behavioral data (cf. Fig.…”

“…From approx. 200 ms on, the RSA patterns also encoded the samples’ numerical magnitude, in terms of a significant numerical distance effect (single-stream: p cluster <0.001; dual-stream: p cluster <0.001), thus replicating and extending previous work (Appelhoff et al, 2022; Luyckx et al, 2019; Sheahan et al, 2021; Spitzer et al, 2017; Teichmann et al, 2018). Descriptively, the numerical distance effect observed in the single-stream task, while robustly significant, appeared weaker than that in the dual-stream task.…”

“…Using electroencephalographic (EEG) recordings and multivariate representational similarity analysis (RSA), recent work has identified a potential neural signature of such psychometric distortions. During processing of symbolic number samples, multivariate EEG patterns have been characterized by a “numerical distance effect”, where the representational similarity of, for instance, “4” and “5” is larger than that between “4” and “6”, which in turn is larger than that between “3” and “7”, and so forth (Appelhoff et al, 2022; Luyckx et al, 2019; Sheahan et al, 2021; Spitzer et al, 2017; Teichmann et al, 2018). Intriguingly, in a multi-sample decision task that promoted anti-compression of number samples in behavior, the “neural numberline” underlying the numerical distance effect was found to be anti-compressed as well (Appelhoff et al, 2022; Luyckx et al, 2019; Spitzer et al, 2017).…”

“…During processing of symbolic number samples, multivariate EEG patterns have been characterized by a “numerical distance effect”, where the representational similarity of, for instance, “4” and “5” is larger than that between “4” and “6”, which in turn is larger than that between “3” and “7”, and so forth (Appelhoff et al, 2022; Luyckx et al, 2019; Sheahan et al, 2021; Spitzer et al, 2017; Teichmann et al, 2018). Intriguingly, in a multi-sample decision task that promoted anti-compression of number samples in behavior, the “neural numberline” underlying the numerical distance effect was found to be anti-compressed as well (Appelhoff et al, 2022; Luyckx et al, 2019; Spitzer et al, 2017). The findings suggested that behaviorally relevant distortions in multi-sample decisions may occur already when the individual samples are being processed.…”