Note that our results can be interpreted as being in conflict with a study by
Konkle, Wang, Hayward, and Moore, (2009).
Konkle et al. (2009) found that motion adaptation aftereffects transfer between vision and touch. Based on this finding, one might expect that, when the surface was seen but not felt (Vision Only condition), any visual adaptation should transfer to the test condition in which only haptic cues were available and no visual information was shown. In the present study we, however, did not find a haptic slant adaptation aftereffect when the surface was presented only visually during adaptation, although it is known that visual slant is subject to visual adaptation (e.g.,
Köhler & Emery, 1947;
Wenderoth, 1970;
Adams, Banks, & van Ee, 2001). There are several possible explanations for this discrepancy. First, we did not measure the visual aftereffects of adaptation and, therefore, can only speculate about a possible existence of such effects in the present task. On the one hand, visual adaptation to slant has been shown to occur for both stereo disparity and perspective cues (e.g.,
Bergman & Gibson, 1959;
Knapen & van Ee, 2006). Thus, we have no reason to believe that the visual system did not adapt in the present case. On the other hand, visual adaptation can be retinal location specific and thus needs fixation rather than free eye movements. Even under strict fixation some adaptation effect might not be complete as microsaccades have been shown to prevent certain types of visual adaptation (e.g.,
Martinez-Conde, Macknik, Troncoso, & Dyar, 2006;
Habtegiorgis, Rifai, Lappe, & Wahl, 2017). In the present study, however, participants were free to make eye movements and it is likely that the participants moved their eyes across the visual scene. This could potentially lead to less adaptation of the visual system. Second, in the study by
Konkle et al. (2009), haptic motion judgments were based on cutaneous cues whereas we used PHANToM force-feedback devices in which only proprioceptive cues were available. Because there is evidence that cutaneous and visual receptive fields overlap in the brain (e.g., see
Graziano & Gross, 1993;
Graziano & Gross, 1995), the processing of these types of information could potentially be more strongly linked. In our case, however, haptic slant estimates were obtained by comparing the finger positions of the left and right hands (see also
Glowania et al., 2020), and thus the estimate with regard to the seen slant of the actual object is indirect at best and more of proprioceptive nature rather than cutaneous. Since in this case the low-level sensations (haptic finger positions and visual object slant) do not directly correspond to the same estimate, this can explain why we do not see any transfer from visual slant to the haptic domain.