Evaluating the concepts of weak and subcognitive integration - an experimental study using vibrotactile sensory augmentation for the blind

Abstract

Navigating in foreign surroundings necessitates peak concentration for blind travellers. Yet, most navigational aids heavily rely on attentional resources as well as on audition. Audition is a modality of supreme importance for the blind, allowing to react to cues of the immediate environment. Thus, it would be highly beneficial for a navigational aid for the blind to not or only partially rely on attentional resources and be easily interpreted and integrated into behaviour. Following the sensorimotor contingency (SMC) theory, which is embedded in the theoretical framework of embodiment, such endeavour has the potential to succeed by employing sensory augmentation devices. According to SMC theory, statistic regularities termed sensorimotor contingencies coupling action and perception are constitutive of conscious perception. Consequentially, since those regularities differ in between modality, also the qualitative experience of different modalities differ. Following this line of thought, new SMCs can be created through sensory augmentation devices and learned by exploring the SMC. The objective of this study is to further investigate if and to what extent such sensory augmentation device can be integrated into behaviour. Therefore, the weak integration hypothesis and the sub-cognitive processing hypothesis as established by Nagel et al. (2005) will be employed to evaluate the integration according to their criteria.Eleven congenitally and adventitiously blind adult subjects were provided with vibrotactile directional information of the magnetic north around the waist through a device termed naviBelt for seven weeks. At the beginning and at the middle of the study the integration of the signal of five participants was assessed using a battery of behavioural tests. These tests consisted of a straight-line-walking task, an angular rotation task and a triangle completion task. Furthermore, throughout the period of study all participants completed preliminary, weekly and final questionnaires, inspired by Kärcher et al. (2012). The questionnaires allowed to gain a more holistic picture of the subjective experience and the self-assessed benefits of the belt. In addition, two deaf-blind participants were provided with the belt for three to four weeks and answered questionnaires adjusted to their needs.The straight-line-walking task showed instant improvements in path stabilization when provided with the belt. In two participants characteristic behaviour of the sub-cognitive processing hypothesis is obtained. An overall improvement independent of whether the belt is worn or not is especially evident after the training period in the angular rotation task. This indicates an enhanced direction estimation accuracy, which is highly related to the understanding of the belt signal. Evidence for enhanced path integration and navigational skills through the belt can be found in the results of the triangle completion task. For two participants the performance improved even with an additional attentional load, hinting towards sub-cognitive processing.Overall, the data supports the weak integration hypothesis and points towards the sub-cognitive processing hypothesis and thus show that SMCs can be learned, which is in line with the theory of embodiment. Crucially, the study further exemplifies how such integration into behaviour can be of great benefit as assistive device for blind and deaf-blind.