Differential callosal contributions to bimanual control in young and older adults

Abstract

Our recent work has shown that older adults are disproportionately impaired at bimanual tasks when the two hands are moving out of phase with each other [Bangert, A. S., Reuter-Lorenz, P. A., Walsh, C. M., Schachter, A. B., & Seidler, R. D. Bimanual coordination and aging: Neurobehavioral implications. Neuropsychologia, 48, 1165-1170, 2010]. Interhemispheric interactions play a key role during such bimanual movements to prevent interference from the opposite hemisphere. Declines in corpus callosum (CC) size and microstructure with advancing age have been well documented, but their contributions to age deficits in bimanual function have not been identified. In the current study, we used structural magnetic resonance and diffusion tensor imaging to investigate age-related changes in the relationships between callosal macrostructure, microstructure, and motor performance on tapping tasks requiring differing degrees of interhemispheric interaction. We found that older adults demonstrated disproportionately poorer performance on out-of-phase bimanual control, replicating our previous results. In addition, older adults had smaller anterior CC size and poorer white matter integrity in the callosal midbody than their younger counterparts. Surprisingly, larger CC size and better integrity of callosal microstructure in regions connecting sensorimotor cortices were associated with poorer motor performance on tasks requiring high levels of interhemispheric interaction in young adults. Conversely, in older adults, better performance on these tasks was associated with larger size and better CC microstructure integrity within the same callosal regions. These findings implicate age-related declines in callosal size and integrity as a key contributor to bimanual control deficits. Further, the differential age-related involvement of transcallosal pathways reported here raises new questions about the role of the CC in bimanual control.

Figure 1

The cross-sectional area of the midsagittal CC was parsed into seven subregions: (1) rostrum, (2) genu, (3) rostral truncus, (4) anterior intermediate truncus, (5) posterior intermediate truncus, (6) isthmus, and (7) splenium.

Figure 2

Bimanual coupling (mean ± 1 SD). Data show an Age × Task interaction with older adults disproportionately impaired at maintaining the appropriate phase delay between hands during out-of-phase tapping. B/w = between; Simul = simultaneous. **p < .01.

Figure 3

(A) Coordinate from Wahl et al. (2007) identified as the average location where fibers connecting hand regions of the primary motor cortex pass through the CC in four representative young adults and (B) four representative older adults. In all participants, this coordinate fell within CC Subregion 5.

Figure 4

CC Subregions 2–7 (mean ± 1 SD). (A) Age × Region interaction; young adults had significantly larger cross-sectional area (normalized to ICA) in CC Subregions 2 and 4. (B) Main effect of age; young adults had significantly greater FA than older adults. (C) Age × Region interaction; young adults had significantly lower radial diffusivity in CC Subregions 3–6. X-sectional = cross-sectional; ICA = intracranial area; FA = fractional anisotropy; rad diff = radial diffusivity; X-sect = cross-sectional. **p < .01; ***p < .001.

Figure 5

(A) Age differences in the relationship between cross-sectional area in CC Subregion 4 (composed of transcallosal fibers connecting SMA) and unimanual tapping. (B) Similar relationships exist between CC4 size and bimanual coupling during out-of-phase tapping. (C) There is no relationship for either age group between CC4 size and bimanual coupling during simultaneous tapping. Solid lines indicate linear regression fit for older adults, whereas dashed lines indicate linear regression fit for young adults. ITI = intertap interval; X-sect = cross-sectional; ICA = intracranial area; B/w = between.

Figure 6

(A) Age differences in the relationship between FA of CC Subregion 6 (composed of transcallosal fibers connecting primary somatosensory cortex) and unimanual tapping. (B) Similar relationships exist between FA in CC6 and bimanual coupling during out-of-phase tapping. (C) There is no relationship for either age group between FA in CC6 and bimanual coupling during simultaneous tapping. Solid lines indicate linear regression fit for older adults, whereas dashed lines indicate linear regression fit for young adults. ITI = intertap interval; B/w = between.

Figure 7

(A) Age differences in the relationship between bimanual coupling during out-of-phase tapping and radial diffusivity in CC Subregion 4 (composed of transcallosal fibers connecting SMA). (B) Similar relationships exist between radial diffusivity in CC Subregion 6 (composed of fibers connecting somatosensory cortex) and bimanual coupling during out-of-phase tapping. (C) There is no relationship for either age group between radial diffusivity in CC4 and bimanual coupling during simultaneous tapping. (D) Nor is there a relationship between bimanual simultaneous tapping and radial diffusivity within CC6 for either age group. Solid lines indicate linear regression fit for older adults, whereas dashed lines indicate linear regression fit for young adults. Rad diff = radial diffusivity; B/w = between.