Anne Hsu

Tuning for spectro-temporal modulations as a mechanism for auditory discrimination of natural sounds

Vocal communicators discriminate conspecific vocalizations from other sounds and recognize the vocalizations of individuals. To identify neural mechanisms for the discrimination of such natural sounds, we compared the linear spectro-temporal tuning properties of auditory midbrain and forebrain neurons in zebra finches with the statistics of natural sounds, including song. Here, we demonstrate that ensembles of auditory neurons are tuned to auditory features that enhance the acoustic differences between classes of natural sounds, and among the songs of individual birds. Tuning specifically avoids the spectro-temporal modulations that are redundant across natural sounds and therefore provide little information; rather, it overlaps with the temporal modulations that differ most across sounds. By comparing the real tuning and a less selective model of spectro-temporal tuning, we found that the real modulation tuning increases the neural discrimination of different sounds. Additionally, auditory neurons discriminate among zebra finch song segments better than among synthetic sound segments.

Modulation Power and Phase Spectrum of Natural Sounds Enhance Neural Encoding Performed by Single Auditory Neurons

We examined the neural encoding of synthetic and natural sounds by single neurons in the auditory system of male zebra finches by estimating the mutual information in the time-varying mean firing rate of the neuronal response. Using a novel parametric method for estimating mutual information with limited data, we tested the hypothesis that song and song-like synthetic sounds would be preferentially encoded relative to other complex, but non-song-like synthetic sounds. To test this hypothesis, we designed two synthetic stimuli: synthetic songs that matched the power of spectral-temporal modulations but lacked the modulation phase structure of zebra finch song and noise with uniform band-limited spectral-temporal modulations. By defining neural selectivity as relative mutual information, we found that the auditory system of songbirds showed selectivity for song-like sounds. This selectivity increased in a hierarchical manner along ascending processing stages in the auditory system. Midbrain neurons responded with highest information rates and efficiency to synthetic songs and thus were selective for the spectral-temporal modulations of song. Primary forebrain neurons showed increased information to zebra finch song and synthetic song equally over noise stimuli. Secondary forebrain neurons responded with the highest information to zebra finch song relative to other stimuli and thus were selective for its specific modulation phase relationships. We also assessed the relative contribution of three response properties to this selectivity: (1) spiking reliability, (2) rate distribution entropy, and (3) bandwidth. We found that rate distribution and bandwidth but not reliability were responsible for the higher average information rates found for song-like sounds.

Methods for the analysis of auditory processing in the brain.

Abstract: Understanding song perception and singing behavior in birds requires the study of auditory processing of complex sounds throughout the avian brain. We can divide the basics of auditory perception into two general processes: (1) encoding, the process whereby sound is transformed into neural activity and (2) decoding, the process whereby patterns of neural activity take on perceptual meaning and therefore guide behavioral responses to sounds. In birdsong research, most studies have focused on the decoding process: What are the responses of the specialized auditory neurons in the song control system? and What do they mean for the bird? Recently, new techniques addressing both encoding and decoding have been developed for use in songbirds. Here, we first describe some powerful methods for analyzing what acoustical aspects of complex sounds like songs are encoded by auditory processing neurons in songbird brain. These methods include the estimation and analysis of spectro‐temporal receptive fields (STRFs) for auditory neurons. Then we discuss the decoding methods that have been used to understand how songbird neurons may discriminate among different songs and other sounds based on mean spike‐count rates.

Persuasive technology for overcoming food cravings and improving snack choices

A central challenge in weight management is the difficulty of overcoming desires for excessive and unhealthy food. Yet, studies show that when people are able to resist their desires for unhealthy choices, they experience pride and satisfaction. In order to alleviate the former and support the latter, we designed, implemented and tested a mobile application for improving snacking behavior. Our application delivers a food craving reduction intervention at the moment of need and allows users to track how often they successfully resisted cravings. Our craving reduction intervention is based on recent research that shows that food cravings can be reduced through imagery techniques. We conducted a week-long evaluation of our application, comparing the effectiveness of our application to a basic tracking application. We found that our imagery application significantly reduced both overall snacking and unhealthy snacking compared to a simple snack-tracking application.

Subjective randomness and natural scene statistics

Accounts of subjective randomness suggest that people consider a stimulus random when they cannot detect any regularities characterizing the structure of that stimulus. We explored the possibility that the regularities people detect are shaped by the statistics of their natural environment. We did this by testing the hypothesis that people’s perception of randomness in two-dimensional binary arrays (images with two levels of intensity) is inversely related to the probability with which the array’s pattern would be encountered in nature. We estimated natural scene probabilities for small binary arrays by tabulating the frequencies with which each pattern of cell values appears. We then conducted an experiment in which we collected human randomness judgments. The results show an inverse relationship between people’s perceived randomness of an array pattern and the probability of the pattern appearing in nature.

Calculating and understanding the value of any type of match evidence when there are potential testing errors

It is well known that Bayes’ theorem (with likelihood ratios) can be used to calculate the impact of evidence, such as a ‘match’ of some feature of a person. Typically the feature of interest is the DNA profile, but the method applies in principle to any feature of a person or object, including not just DNA, fingerprints, or footprints, but also more basic features such as skin colour, height, hair colour or even name. Notwithstanding concerns about the extensiveness of databases of such features, a serious challenge to the use of Bayes in such legal contexts is that its standard formulaic representations are not readily understandable to non-statisticians. Attempts to get round this problem usually involve representations based around some variation of an event tree. While this approach works well in explaining the most trivial instance of Bayes’ theorem (involving a single hypothesis and a single piece of evidence) it does not scale up to realistic situations. In particular, even with a single piece of match evidence, if we wish to incorporate the possibility that there are potential errors (both false positives and false negatives) introduced at any stage in the investigative process, matters become very complex. As a result we have observed expert witnesses (in different areas of speciality) routinely ignore the possibility of errors when presenting their evidence. To counter this, we produce what we believe is the first full probabilistic solution of the simple case of generic match evidence incorporating both classes of testing errors. Unfortunately, the resultant event tree solution is too complex for intuitive comprehension. And, crucially, the event tree also fails to represent the causal information that underpins the argument. In contrast, we also present a simple-to-construct graphical Bayesian Network (BN) solution that automatically performs the calculations and may also be intuitively simpler to understand. Although there have been multiple previous applications of BNs for analysing forensic evidence—including very detailed models for the DNA matching problem, these models have not widely penetrated the expert witness community. Nor have they addressed the basic generic match problem incorporating the two types of testing error. Hence we believe our basic BN solution provides an important mechanism for convincing experts—and eventually the legal community—that it is possible to rigorously analyse and communicate the full impact of match evidence on a case, in the presence of possible errors.

Designing for Psychological Change: Individuals’ Reward and Cost Valuations in Weight Management

Background Knowledge of the psychological constructs that underlie behavior offers valuable design opportunities for persuasive systems. We use the decision theory, which describes how behavior is underpinned by reward-cost valuations, as a framework for investigating such psychological constructs to deliver design objectives for weight management technologies. Objective We applied a decision theory–based analysis in the domain of weight management to understand the rewards and costs that surround individuals’ weight management behaviors, with the aim of uncovering design opportunities for weight management technologies. Methods We conducted qualitative interviews with 15 participants who were or had been trying to lose weight. Thematic analysis was used to extract themes that covered the rewards and costs surrounding weight management behaviors. We supplemented our qualitative study with a quantitative survey of 100 respondents investigating the extent to which they agreed with statements reflecting themes from the qualitative study. Results The primary obstacles to weight management were the rewards associated with unhealthy choices, such as the pleasures of unhealthy foods and unrestricted consumption in social situations, and the significant efforts required to change habits, plan, and exercise. Psychological constructs that supported positive weight management included feeling good after making healthy choices, being good to oneself, experiencing healthy yet still delicious foods, and receiving social support and encouraging messages (although opinions about encouraging messages was mixed). Conclusions A rewards-costs driven enquiry revealed a wide range of psychological constructs that contribute to discouraging and supporting weight management. The constructs extracted from our qualitative study were verified by our quantitative survey, in which the majority of respondents also reported similar thoughts and feelings. This understanding of the rewards and costs surrounding weight management offers a range of new opportunities for the design of weight management technologies that enhance the encouraging factors and alleviate the discouraging ones.

When Absence of Evidence Is Evidence of Absence: Rational Inferences From Absent Data

Identifying patterns in the world requires noticing not only unusual occurrences, but also unusual absences. We examined how people learn from absences, manipulating the extent to which an absence is expected. People can make two types of inferences from the absence of an event: either the event is possible but has not yet occurred, or the event never occurs. A rational analysis using Bayesian inference predicts that inferences from absent data should depend on how much the absence is expected to occur, with less probable absences being more salient. We tested this prediction in two experiments in which we elicited peoples judgments about patterns in the data as a function of absence salience. We found that people were able to decide that absences either were mere coincidences or were indicative of a significant pattern in the data in a manner that was consistent with predictions of a simple Bayesian model.

Exploring Human Cognition Using Large Image Databases

Most cognitive psychology experiments evaluate models of human cognition using a relatively small, well-controlled set of stimuli. This approach stands in contrast to current work in neuroscience, perception, and computer vision, which have begun to focus on using large databases of natural images. We argue that natural images provide a powerful tool for characterizing the statistical environment in which people operate, for better evaluating psychological theories, and for bringing the insights of cognitive science closer to real applications. We discuss how some of the challenges of using natural images as stimuli in experiments can be addressed through increased sample sizes, using representations from computer vision, and developing new experimental methods. Finally, we illustrate these points by summarizing recent work using large image databases to explore questions about human cognition in four different domains: modeling subjective randomness, defining a quantitative measure of representativeness, identifying prior knowledge used in word learning, and determining the structure of natural categories.

Sampling Assumptions Affect Use of Indirect Negative Evidence in Language Learning.

A classic debate in cognitive science revolves around understanding how children learn complex linguistic patterns, such as restrictions on verb alternations and contractions, without negative evidence. Recently, probabilistic models of language learning have been applied to this problem, framing it as a statistical inference from a random sample of sentences. These probabilistic models predict that learners should be sensitive to the way in which sentences are sampled. There are two main types of sampling assumptions that can operate in language learning: strong and weak sampling. Strong sampling, as assumed by probabilistic models, assumes the learning input is drawn from a distribution of grammatical samples from the underlying language and aims to learn this distribution. Thus, under strong sampling, the absence of a sentence construction from the input provides evidence that it has low or zero probability of grammaticality. Weak sampling does not make assumptions about the distribution from which the input is drawn, and thus the absence of a construction from the input as not used as evidence of its ungrammaticality. We demonstrate in a series of artificial language learning experiments that adults can produce behavior consistent with both sets of sampling assumptions, depending on how the learning problem is presented. These results suggest that people use information about the way in which linguistic input is sampled to guide their learning.