Spatial-numerical associations: Shared symbolic and non-symbolic numerical representations

Area of Science:

• Cognitive Neuroscience
• Psychology
• Numerical Cognition

Background:

• Humans and animals possess an approximate number system (ANS) for quantity discrimination, independent of counting.
• A prominent theory posits shared, spatial number representations for both symbolic (e.g., digits) and non-symbolic (e.g., dot arrays) numerical processing.
• An alternative view suggests quantity-related representations without specific number representations, implying separate processing for symbolic and non-symbolic numbers.

Purpose of the Study:

• To review and analyze behavioral evidence concerning spatial biases in numerical processing.
• To evaluate whether these biases support a shared representation model or distinct representation models of the ANS.
• To propose a resolution for the current theoretical impasse regarding number representations.

Main Methods:

• Review of studies examining spatial biases in non-symbolic numerical tasks.
• Review of studies examining spatial biases in symbolic numerical tasks, including the spatial-numerical association of response codes (SNARC) effect.
• Discussion of integrating behavioral findings with neuroimaging data.

Main Results:

• Behavioral evidence regarding spatial biases is currently divided.
• Bisection tasks suggest support for shared representations.
• SNARC effect studies currently favor distinct representations.

Conclusions:

• The existing behavioral data is insufficient to definitively distinguish between shared and distinct number representation theories.
• Combining behavioral paradigms (e.g., bisection, SNARC) with neuroimaging is crucial for resolving the debate.
• Future research integrating these approaches may clarify the nature of numerical representations in the ANS.