Numerical distance effects in visual search

Abstract

We present three experiments in which observers searched for a target digit among distractor digits in displays in which the mean numerical target-distractor distance was varied. Search speed and accuracy increased with numerical distance in both target-present and target-absent trials (Exp. 1A). In Experiment 1B, the target 5 was replaced with the letter S. The results suggest that the findings of Experiment 1A do not simply reflect the fact that digits that were numerically closer to the target coincidentally also shared more physical features with it. In Experiment 2, the numerical distance effect increased with set size in both target-present and target-absent trials. These findings are consistent with the view that increasing numerical target-distractor distance affords faster nontarget rejection and target identification times. Recent neurobiological findings (e.g., Nieder, 2011) on the neuronal coding of numerosity have reported a width of tuning curves of numerosity-selective neurons that suggests graded, distance-dependent coactivation of the representations of adjacent numbers, which in visual search would make it harder to reject numerically closer distractors as nontargets.

The number–weight illusion

“…From the perspective of numerical cognition, our results add to the list of findings that numerical magnitude not only affects purely cognitive judgments about the abstract order relations of numbers but extends to more versatile tasks in which motor efference copies are compared to sensory feedback. Our findings thus complement and extend earlier studies demonstrating that symbolic numerical magnitude systematically influences complex activities such as visual search (Schwarz & Eiselt, 2012;Sobel, Puri, & Faulkenberry, 2016), the scaling of manual force in motor commands (Fischer & Miller, 2008;Lindemann et al, 2007;, or the evaluation of the temporal order of events (Gevers et al, 2003;Müller & Schwarz, 2008;Schwarz & Eiselt, 2009). In contrast to these findings the present study is the first to report the a priori seemingly implausible finding shown in Fig.…”

Interaction between numbers and size during visual search

“…Furthermore, Schwarz and Eiselt ( 2012 ) recently demonstrated that the magnitude information of different simultaneously presented Arabic digits becomes automatically activated and affects the visual search for a target number in these displays. The authors required their participants to find a target digit among distractor digits in displays in which the average numerical distance between the target and distractors was systematically varied.…”

Using multidimensional scaling to quantify similarity in visual search and beyond

“…Numerical meaning is extracted quickly from visual displays (Corbett, Oriet, & Rensink, 2006). Indeed, Schwarz and Eiselt (2012) found that the presence of numerically similar distractors (e.g., 4, 6) decreased speed and accuracy in search for the number 5 compared with dissimilar distractors (e.g., 1, 9), suggesting that attention was drawn to digits that were numerically close to the target value. They did not find this pattern of results when participants searched for the letter S, even though it is highly visually similar to the digit 5, suggesting that the visual characteristics of the numbers were unlikely to be driving their semantic effects.…”

“…Godwin, Hout, and Menneer (2014) built upon the study by Schwarz and Eiselt (2012) by quantifying visual similarity using MDS ratings of the digits, and by tracking the oculomotor behavior of our participants. One group of participants provided visual similarity ratings for the numbers 0–9; we then analyzed the data using MDS, and indexed visual similarity via the distance between each pair of numbers in MDS space.…”

“…One group of participants provided visual similarity ratings for the numbers 0–9; we then analyzed the data using MDS, and indexed visual similarity via the distance between each pair of numbers in MDS space. Semantic similarity (as in the Schwarz and Eiselt, 2012, study) was indexed by the numerical distance between the numbers. One potential concern with indexing visual similarity using MDS measures is that – despite our instructions to rate the stimuli purely based on their visual characteristics – it is possible that the semantic (numerical) similarity of the numbers also affected the ratings provided by our participants.…”