The Collaboration between Hearing Aid Users and Artificial Intelligence to Optimize Sound.

Hearing aid gain and signal processing are based on assumptions about the average user in the average listening environment, but problems may arise when the individual hearing aid user differs from these assumptions in general or specific ways. This article describes how an artificial intelligence (AI) mechanism that operates continuously on input from the user may alleviate such problems by using a type of machine learning known as Bayesian optimization. The basic AI mechanism is described, and studies showing its effects both in the laboratory and in the field are summarized. A crucial fact about the use of this AI is that it generates large amounts of user data that serve as input for scientific understanding as well as for the development of hearing aids and hearing care. Analyses of users' listening environments based on these data show the distribution of activities and intentions in situations where hearing is challenging. Finally, this article demonstrates how further AI-based analyses of the data can drive development.

Effects of Surprisal and Locality on Danish Sentence Processing: An Eye-Tracking Investigation.

An eye-tracking experiment in Danish investigates two dominant accounts of sentence processing: locality-based theories that predict a processing advantage for sentences where the distance between the major syntactic heads is minimized, and the surprisal theory which predicts that processing time increases with big changes in the relative entropy of possible parses, sometimes leading to anti-locality effects. We consider both lexicalised surprisal, expressed in conditional trigram probabilities, and syntactic surprisal expressed in the manipulation of the expectedness of the second NP in Danish constructions with two postverbal NP-objects. An eye-tracking experiment showed a clear advantage for local syntactic relations, with only a marginal effect of lexicalised surprisal and no effect of syntactic surprisal. We conclude that surprisal has a relatively marginal effect, which may be clearest for verbs in verb-final languages, while locality is a robust predictor of sentence processing.

Investigating lexical competition and the cost of phonemic restoration.

Due to phonemic restoration, listeners can reliably perceive words when a phoneme is replaced with noise. The cost associated with this process was investigated along with the effect of lexical uniqueness on phonemic restoration, using data from a lexical decision experiment where noise replaced phonemes that were either uniqueness points (the phoneme at which a word deviates from all nonrelated words that share the same onset) or phonemes immediately prior to these. A baseline condition was also included with no noise-interrupted stimuli. Results showed a significant cost of phonemic restoration, with 100 ms longer word identification times and a 14% decrease in word identification accuracy for interrupted stimuli compared to the baseline. Regression analysis of response times from the interrupted conditions showed no effect of whether the interrupted phoneme was a uniqueness point, but significant effects for several temporal attributes of the stimuli, including the duration and position of the interrupted segment. These results indicate that uniqueness points are not distinct breakpoints in the cohort reduction that occurs during lexical processing, but that temporal properties of the interrupted stimuli are central to auditory word recognition. These results are interpreted in the context of models of speech perception.

Cognitive Load in Voice Therapy Carry-Over Exercises.

Purpose: The cognitive load generated by online speech production may vary with the nature of the speech task. This article examines 3 speech tasks used in voice therapy carry-over exercises, in which a patient is required to adopt and automatize new voice behaviors, ultimately in daily spontaneous communication. Method: Twelve subjects produced speech in 3 conditions: rote speech (weekdays), sentences in a set form, and semispontaneous speech. Subjects simultaneously performed a secondary visual discrimination task for which response times were measured. On completion of each speech task, subjects rated their experience on a questionnaire. Results: Response times from the secondary, visual task were found to be shortest for the rote speech, longer for the semispontaneous speech, and longest for the sentences within the set framework. Principal components derived from the subjective ratings were found to be linked to response times on the secondary visual task. Acoustic measures reflecting fundamental frequency distribution and vocal fold compression varied across the speech tasks. Conclusions: The results indicate that consideration should be given to the selection of speech tasks during the process leading to automation of revised speech behavior and that self-reports may be a reliable index of cognitive load.

Givenness, complexity, and the Danish dative alternation.

In this study, we investigated the interaction between givenness and complexity on the choice of syntactic structure, via two experiments using speeded acceptability judgments. Experiment 1 showed that for the Danish dative alternation, given-new orders are only easier to process for double-object or NP constructions, whereas PP constructions are unaffected. This replicates previous findings for the English dative alternation. Experiment 2 revealed that when a long NP precedes a short NP-a suboptimal complexity relation-the effect of givenness is neutralized, whereas givenness remains influential when the complexity relation between the NPs in the sentence is optimal. This is consistent with the view that in online parsing, the actual syntactic structure-building process is primary, whereas any higher-order computations such as discourse linking are secondary. The relative complexity of the NPs in the double-object construction directly affects the structure-building process, whereas the decoding of the discourse structure is a later and less crucial phenomenon, resulting in neutralization of the givenness effect in cases in which the complexity relation is suboptimal.

Probability and surprisal in auditory comprehension of morphologically complex words.

Two auditory lexical decision experiments document for morphologically complex words two points at which the probability of a target word given the evidence shifts dramatically. The first point is reached when morphologically unrelated competitors are no longer compatible with the evidence. Adapting terminology from Marslen-Wilson (1984), we refer to this as the word's initial uniqueness point (UP1). The second point is the complex uniqueness point (CUP) introduced by Balling and Baayen (2008), at which morphologically related competitors become incompatible with the input. Later initial as well as complex uniqueness points predict longer response latencies. We argue that the effects of these uniqueness points arise due to the large surprisal (Levy, 2008) carried by the phonemes at these uniqueness points, and provide independent evidence that how cumulative surprisal builds up in the course of the word co-determines response latencies. The presence of effects of surprisal, both at the initial uniqueness point of complex words, and cumulatively throughout the word, challenges the Shortlist B model of Norris and McQueen (2008), and suggests that a Bayesian approach to auditory comprehension requires complementation from information theory in order to do justice to the cognitive cost of updating probability distributions over lexical candidates.