Stroop and Garner effects in comparative judgment of numerals: The role of attention.

Abstract

In 7 experiments, participants selected the larger member of pairs of digits that differed in numerical magnitude as well as in physical size. Selective attention to the relevant dimension (number or size) was gauged by Garner and Stroop interference, both of which varied considerably as a function of the number and relative discriminability of values along the constituent dimensions. When the to-be-ignored dimension was more discriminable, sizable Garner and Stroop effects affected performance on the relevant dimension. When it was less discriminable, Garner and Stroop effects were considerably smaller regardless of whether the relevant dimension was numerical or physical size. The sensitivity of Stroop interference to manipulations of discriminability is accounted for by the allocation of attention to the constituent dimensions. The demonstrated malleability of the Stroop effect is incompatible with claims of strong automaticity in numerical processing.

Are numbers special? An overview of chronometric, neuroimaging, developmental and comparative studies of magnitude representation

“…A more convincing finding would be that numerical and non-numerical quantities interact with each other. This is exactly what is found with the size congruity paradigm (Algom et al, 1996;Cohen Kadosh and Henik, 2006;Cohen Kadosh et al, 2007e;Fias et al, 2002;Henik and Tzelgov, 1982;Hurewitz et al, 2006;Pansky and Algom, 1999;Schwarz and Ischebeck, 2003;Tzelgov et al, 1992). In this Stroop-like task, numerical and physical dimensions are varied independently.…”

“…As previously mentioned in the behavioral section of this review, a different task that might be used to examine the uniqueness of numbers is the size congruity task (Cohen Kadosh and Henik, 2006;Cohen Kadosh et al, 2007e;Henik and Tzelgov, 1982;Pansky and Algom, 1999;Schwarz and Ischebeck, 2003;Tzelgov et al, 1992). The slowing down of responses (i.e., the size congruity effect) by the irrelevant dimension suggests that processing of numerical and physical dimensions overlap up to a certain point, when selection of the relevant dimension occurs.…”

“…However, a possible explanation for this exclusive numerical activation can be found in the automaticity of number processing. Namely, the magnitude of numbers is automatically processed even when not relevant to the task, as indicated, for example, by the size congruity effect (Cohen Kadosh and Henik, 2006;Cohen Kadosh et al, 2007e;Henik and Tzelgov, 1982;Pansky and Algom, 1999;Schwarz and Ischebeck, 2003;Tzelgov et al, 1992). In contrast, the degree of automaticity for letters is much smaller than for numbers (Gevers et al, 2003).…”

Extracting Meaning from Sound: Nomic Mappings, Everyday Listening, and Perceiving Object Size from Frequency

“…Second, the numerical values chosen were scaled versions of actual bar lengths that, when struck, would produce the particular frequencies. Third, Pansky and Algom (1999) used the Garner interference paradigm and demonstrated that numerical magnitude and physical size are perceptually integral.…”

Implicit Spatial Length Modulates Time Estimates, But Not Vice Versa

“…Tests of cross-dimensional relationships often manipulate more levels of one dimension than of the other, creating an imbalance in discriminability (see Pansky & Algom, 1999). In the space-time experiments by Casasanto & Boroditsky (2008), however, there were 9 levels of each dimension fully crossed, to equate discriminability.…”