The Limits of Touch: Spatial acuity for frequency-resolved air-borne ultrasound vibrotactile stimuli

Spatial acuity is a fundamental property of any sensory system. In the case of the somatosensory system, the two-point discrimination (2PD) test has long been used to investigate the spatial resolution of tactile perception. The somatosensory system comprises multiple mechanoreceptive channels, each tuned to specific vibrotactile frequencies. In particular, the rapidly adapting channel (RA) responds to low-frequency vibration and is thought to have high spatial acuity. The Pacinian channel (PC) responds to high-frequency vibration and is thought to convey little or no spatial information. However, the mechanical stimulations used in most 2PD tests make it difficult to disentangle the contribution of each mechanoreceptive channel to spatial tactile perception. Here we developed a novel 2PD test based on ultrasound stimulation to deliver frequency-resolved vibrotactile stimuli designed to preferentially activate specific tactile channels. Across four experiments, we systematically investigated the spatial resolution of the two main vibrotactile channels. Contrary to the textbook view of poor spatial resolution for PC-like stimuli, we found that high-frequency vibration produced surprisingly good spatial acuity. This effect remained after controlling for differences between the channels in stimulus detectability and perceived intensity. Laser doppler vibrometry experiments confirmed that the acuity of the PC channel was not simply an artifact of the skin’s resonance to high-frequency mechanical stimulation. Thus, PC receptors may transmit substantial spatial information, despite their sparse distribution, deep location, and large receptive fields.

Experiment 3 also allowed to directly compare the 2PD thresholds in the physically showing a significant difference (t 19 = -2.41; p = .013; d z = -.540). That is, when the 2 4 2 intensity of the 200 Hz stimulus was reduced, in our case by around 50 %, so as to match 2 4 3 the 50 Hz stimulus, the participants' 2PD threshold was significantly higher (i.e., worse 2 4 4 acuity). This suggests that spatial acuity depends on stimulus intensity.

4 5
Finally, to investigate whether the effect of frequency and perceived intensity on 2 4 6 spatial acuity was mediated by the detectability of the stimuli, we ran a one-way repeated participants' response bias (C) was non-significant (F 2,38 = 2.40; p = .104; η 2 p = 112). significantly different. This result suggests that the relationship between 2 6 0 physical/perceived intensity and spatial acuity is not simply explained by higher 2 6 1 detectability. In fact, even when the stimuli are equally detectable, the perceived intensity 2 6 2 of the stimuli still influences spatial acuity.  non-neural effect due to skin mechanics. We therefore used LDV to measure the actual 3 0 3 peak-to-peak displacement of the skin caused by 50 Hz ultrasound stimuli and by Contrary to the differential mechanical resonance hypothesis, we found that in all 3 0 6 the four participants tested, the average peak-to-peak skin displacement was in fact greater  Next, we quantified the average skin displacement due to a 200 Hz stimulus that resonance of the skin to 50 Hz stimulation. Rather, they suggest a frequency-dependent 3 1 7 non-linear relationship between physical skin displacement and perceived intensity. stimuli. Contrary to a differential mechanical resonance hypothesis, we found that the average peak-to-peak skin Next, we tested whether the effects of frequency and intensity on spatial acuity 3 2 8 might be due to the interference created by the two mechanical propagation waves on the 3 2 9 skin that inevitably arise when two points are stimulated simultaneously 24,25 . The 2D and 1D plots of the skin displacement induced by the two points at the participants' 2PD (4.51 ± 3.20 µ m) ( Figure 5 and Movie S1), thus refuting the idea that the observed 2PD propagations. These data also confirm the success of our attempts to design ultrasound 3 3 7 stimuli with minimal side-lobes for the specific purpose of acuity testing (see Methods). Spatial acuity is a fundamental property of the somatosensory system, but it has been 3 4 6 studied using exclusively mechanical pressure stimuli. This means that spatial acuity has been 3 4 7 properly characterised only for one of the many mechanoreceptive channels constituting the was well below the RA/Meissner channel threshold, they assumed that any resulting percept was 3 8 2 due to the PC channel. They showed that a perceptual experience could be evoked from the PC 3 8 3 channel for pulses at frequencies down to 6 Hz. Moreover, these stimuli produced an experience projections, and instead suggest a common perceptual dimension of frequency that is encoded by single receptor type and did not explore the frequency continuum. In addition, we have focussed 3 9 0 on the spatial percepts generated by frequency-targeted stimuli, in contrast to the more extensive 3 9 1 literature on temporal aspects of perception. However, both sets of results indicate that the role 3 9 2 of the PC channel in human somatosensory perception may be wider than previously thought, 3 9 3 given that PCs encode a surprising amount of temporal information 36 and also spatial 3 9 4 information (present results).

9 5
Our results have special implications for haptic applications. Mid-air haptics using 3 9 6 ultrasound stimuli to produce tactile experiences without mechanical contact between skin and 3 9 7 stimulator. These devices cannot provide steady pressure or low-frequency stimulation, and 3 9 8 therefore cannot currently produce experiences such as hand-object interactions, grasping etc.

9 9
For this reason, recent interest has often focussed on temporal pattern perception, including were excluded because of difficulty in detecting some stimuli, particularly at lower frequencies.

0 9
Other ultrasound tactile studies report similar exclusions 38 . As a consequence of these relation between stimulus intensity and spatial acuity is not well understood. Our own data show 4 1 2 a relation between sensory detection, which depends mainly on stimulus intensity, and spatial 4 1 3 acuity. We therefore speculate that stronger stimulation might reveal even better spatial acuity 4 1 4 of the PC channel. On the other hand, several studies suggest that spatial information and 4 1 5 force/pressure information are processed in separate brain pathways 39,40 .

1 6
Second, ultrasonic stimulation provides a focal point of stimulation that is much larger 4 1 7 than the probes conventionally used in acuity testing. Our LVD recordings indicate that each participants.

8 3
The staircase procedures resulted in two-point discrimination thresholds (2PDT), and were corresponds to better spatial resolution. Error bars represent the SEM. descending and ascending staircases, respectively. B. Two-point discrimination thresholds.

9 9
Red and green lines show participants' responses to descending and ascending staircases, Experiments 1 and 2 were successfully replicated. The 200 Hz stimulus produced lower 2PD 6 0 2 thresholds (i.e., better spatial resolution) than the 50 Hz stimulus when the two conditions had Response bias data. The second part of Experiment 3 investigated whether the effect of 6 0 6 frequency and perceived intensity on spatial acuity was mediated by the detectability of the 6 0 7 stimuli. The data on the participants' response bias were non-significant. D. Sensitivity data.

0 8
Participants' sensitivity to the 50 Hz stimulus was significantly lower than both the physically three ultrasound stimuli. Contrary to a differential mechanical resonance hypothesis, we found 6 1 2 that the average peak-to-peak skin displacement was greater in the 50 Hz condition, compared to stimuli that were matched for physical intensity, and that produced much smaller skin conditions. We LDV recording in each stimulation condition at its 2PD threshold for a 6 1 9 representative participant. B. Cross-section between the two maxima from the same data 6 2 0 shown in A. (See also Video S1).