Understanding Estimations of Magnitudes: An fMRI Investigation

Sarit Ashkenazi, Yarden Gliksman, Avishai Henik

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


The current study examined whether discrete numerical estimation is based on the same cognitive process as estimation of continuous magnitudes such as weight and time. While the verbal estimation of numerical quantities has a contingent unit of measurement (e.g., how many cookies fit in a cookie jar? _X_ cookies), estimation of time and weight does not (e.g., how much time does it take to fill a bath with water? _X_ minutes/hours/seconds). Therefore, estimation of the latter categories has another level of difficulty, requiring extensive involvement of cognitive control. During a functional magnetic resonance imaging (fMRI) scan, 18 students performed estimations with three estimation categories: number, time, and weight. Estimations elicited activity in multiple brain re-gions, mainly: (1) visual regions including bilateral lingual gyrus), (2) parietal regions including the left angular gyrus and right supramarginal gyrus, and (3) the frontal regions (cingulate gyrus and the inferior frontal cortex). Continuous magnitude estimations (mostly time) produced different frontal activity than discrete numerical estimations did, demonstrating different profiles of brain activations between discrete numerical estimations and estimations of continuous magnitudes. The activity level in the right middle and inferior frontal gyrus correlated with the tendency to give extreme responses, signifying the importance of the right prefrontal lobe in estimations.

Original languageEnglish
Article number104
JournalBrain Sciences
Issue number1
StatePublished - 1 Jan 2022


  • Angular gyrus
  • Cognitive estimation
  • Executive functions
  • Numerical estimation

ASJC Scopus subject areas

  • General Neuroscience


Dive into the research topics of 'Understanding Estimations of Magnitudes: An fMRI Investigation'. Together they form a unique fingerprint.

Cite this