Sensory substitution refers to the concept of feeding information to the brain via an atypical sensory pathway. We here examined the degree to which participants (deaf and hard of hearing) can learn to identify sounds that are algorithmically translated into spatiotemporal patterns of vibration on the skin of the wrist. In a three-alternative forced choice task, participants could determine the identity of up to 95% and on average 70% of the stimuli simply by the spatial pattern of vibrations on the skin. Performance improved significantly over the course of 1 month. Younger participants tended to score better, possibly because of higher brain plasticity, more sensitive skin, or better skills at playing digital games. Similar results were obtained with pattern discrimination, in which a pattern representing the sound of one word was presented to the skin, followed by that of a second word. Participants answered whether the word was the same or different. With minimal difference pairs (distinguished by only one phoneme, such as “house” and “mouse”), the best performance was 83% (average of 62%), while with non-minimal pairs (such as “house” and “zip”) the best performance was 100% (average of 70%). Collectively, these results demonstrate that participants are capable of using the channel of the skin to interpret auditory stimuli, opening the way for low-cost, wearable sensory substitution for the deaf and hard of hearing communities.

感觉替代是指通过非典型感觉途径向大脑提供信息的概念。我们在这里检查了参与者（聋哑人和听力障碍者）可以学会识别声音的程度，这些声音在算法上被转换为手腕皮肤上的时空振动模式。在三项强制选择任务中，参与者可以仅通过皮肤上振动的空间模式来确定多达95％的刺激身份，并且平均可以确定70％的刺激身份。在1个月的时间内，性能有了显着提高。年轻的参与者往往得分更高，这可能是因为大脑的可塑性更高，皮肤更敏感或玩数字游戏的技能更高。模式识别获得了类似的结果，在皮肤上呈现出代表一个单词的声音的图案，然后是第二个单词的声音。与会者回答了这个词是相同还是不同。使用最小差异对（只有一个音素（例如“ house”和“ mouse”）区分时，最佳性能为83％（平均为62％），而使用非最小对（例如“ house”和“ zip”） ”）的最佳效果是100％（平均值为70％）。总的来说，这些结果表明，参与者能够使用皮肤的通道来解释听觉刺激，从而为低成本，可穿戴的感觉替代提供了途径，以替代聋人和重听者。例如“ house”和“ mouse”），最佳性能为83％（平均值为62％），而使用非最小对（例如“ house”和“ zip”），则最佳性能为100％（平均值为70％）。总的来说，这些结果表明，参与者能够使用皮肤的通道来解释听觉刺激，从而为低成本，可穿戴的感觉替代提供了途径，以替代聋人和重听者。例如“ house”和“ mouse”），最佳性能为83％（平均值为62％），而使用非最小对（例如“ house”和“ zip”），则最佳性能为100％（平均值为70％）。总的来说，这些结果表明，参与者能够使用皮肤的通道来解释听觉刺激，从而为低成本，可穿戴的感觉替代提供了途径，以替代聋人和重听者。