As noted in the previous post, there is a right brain hemisphere dominance in attending to visuospatial and numerical information. (Rogers et al 2013) So when patients with a right parietal lesion and therefore a spatial deficit for left side stimuli, are asked to point to the mid point of a line they point to a position right of centre. Or if asked for a number halfway between 2 and 6 might reply 5. (Zorz1 et al 2002)

Subjects without this deficit, judge the centre of visually presented horizontal lines fairly accurately, only making small errors to the left for short lines and small errors to the right for larger lines (Brooks, Della Sala, & Darling, 2014)

Things get interesting when the line to be bisected is made of repeated arabic numerals or number words. If the line is made up of small numbers the perceived midpoint shifts to the left and if the line is made up of large numbers shifts to the right. (Calabria & Rossetti, 2005; Fischer, 2001).

Care in avoiding potential methodological problems such as getting matching line lengths and equal image densities were explored by using french number words of the same length, similar density but different magnitudes; in the case below DEUX and NEUF. (Calabria and Rossetti 2005)

The left edges of the two lines are both defined by a solid vertical edge and in the top pair the right hand edge by two points. But in the canonical and mirror presentations the characters are shifted so that the lines both start and end with the same letter ‘E’.

Importantly subjects were requested to bisect the strings without any reference being made to the numbers making up the lines.

Discussion

Although in the literature this effect is described as a bias, I think that it demonstrates a priming effect operating on the spatial representation of number. This is because whilst the attention of subjects is not drawn to the number words that form the lines, the words do in fact subliminally effect the decisions being made by the subjects.

The arguments of Calabria and Rossetti are more nuanced than I have presented, including the fact that drawing attention to the task irrelevant number elicits a stronger effect. In most related experiments this is done by asking if the number displayed is odd or even.

However a great deal of unconscious thinking is going on in identifying and extracting the number words from the lines, so one might expect that lines made up of number dots would work.

Bibliography

Brooks, J.L., Della Sala, S. & Darling, S. (2014) Representational Pseudoneglect: A Review. Neuropsychology review, vol 24, no. 2, pp. 148-165

Calabria, M., and Rossetti, Y. (2005) Interference between number processing and line bisection: a methodology. Neuropsychologia 43 779–78

Fischer, M. H. (2001). Number processing induces spatial performance biases. Neurology, 822–826.

Rogers, L. J., Vallortigara G. , R. J. Andrew R. J. (2013) Divided Brains Cambridge Univ. Press, New York.

Shaki S., Fischer M., & Petrusic W. (2009). Reading habits for both words and numbers contribute to the SNARC effect Psychonomic Bulletin & Review, 16 (2), 328-331

Zorzi M., Priftis K. and Umila C. (2002) Brain Damage: Neglect disrupts the mental number line. NATURE | VOL 417 | 9 MAY 2002 | 138-139