![Professor Jeffrey Bowers](https://research-information.bris.ac.uk/ws/files/311210978/114916.jpg)
Professor Jeffrey Bowers
B.Sc.(Tor.), Ph.D.(Arizona)
Current positions
Professor
School of Psychological Science
Contact
Press and media
Many of our academics speak to the media as experts in their field of research. If you are a journalist, please contact the University’s Media and PR Team:
Research interests
My research addresses a range of issues in language and memory. In one line of work I have attempted to gain insight into how word knowledge is coded in the brain. On one general view, word knowledge (and indeed all forms of knowledge) is coded in a distributed (and non-symbolic) manner, such that a word is coded as a pattern of activation across a set of units (neurons), with no one unit devoted to a single letter or word (typically associated with the PDP approach). On another view, word knowledge is coded in a localist (and symbolic) manner, with each letter and word uniquely coded by an individual unit. I’ve carried out a series of behavioral experiments that provide evidence that letters and words are coded in a localist and symbolic manner (e.g., Davis & Bowers, 2005, 2006), as well as some computer simulations that support this conclusion (Bowers, Damian, & Davis, in press, Psychological Review, Bowers & Davis, 2009). I’ve also argued that localist models are more biological plausible than the distributed representations learned in PDP networks (Bowers, 2009).
Another line of research attempts to further our understanding of the learning mechanisms that support written and spoken word perception. In one study we have provided evidence that the age at which a word is learned is as important as the frequency with a word is practiced (Stadthagen-Gonzalez et al., 2004). At the same time, we have provided evidence that early learning leaves an indelible imprint on our ability to perceive the sounds of a language (Bowers, Mattys, and Gage, 2009). In this project, we found that persons who were exposed to Zulu and Hindi early in life could relearn phoneme contrasts in these languages following years of isolation from Zulu or Hindi. By contrasts, adults who were never exposed to these languages as children could not learn the contrasts. That is, early exposure to the phonemes in a language is special. In another recent project, we have provided evidence that word learning involves a consolidation process, in which learning improves over time (perhaps during sleep) in the absence of further training (Clay et al., 2007).
Projects and supervisions
Research projects
Exploring the multiple loci of learning and computation in simple artificial neural networks
Principal Investigator
Managing organisational unit
School of Psychological ScienceDates
01/03/2023 to 31/08/2024
M and M
Principal Investigator
Description
Can we use Deep Neural Networks to understand how the mind works? The 5-year ERC grant entitled “Generalisation in Mind and Machine” compares how humans and artificial neural networks generalise…Managing organisational unit
School of Psychological ScienceDates
01/09/2017 to 31/08/2022
ARE FORM-RELATED WORDS CO-ACTIVATED DURING THE GENERATION OF A SPOKEN RESPONSE?
Principal Investigator
Managing organisational unit
School of Psychological ScienceDates
01/06/2006 to 01/06/2007
THE IMPACT OF EARLY EXPERIENCE ON LATER SPEECH PERCEPTION
Principal Investigator
Managing organisational unit
School of Psychological ScienceDates
05/09/2005 to 05/03/2009
IS SPEECH PERCEPTION INFLUENCED BY TOP-DOWN FEEDBACK
Principal Investigator
Managing organisational unit
School of Psychological ScienceDates
01/07/2005 to 01/07/2006
Thesis supervisions
Publications
Recent publications
02/06/2023Convolutional Neural Networks Trained to Identify Words Provide a Surprisingly Good Account of Visual Form Priming Effects
Computational Brain & Behavior
Mixed Evidence for Gestalt Grouping in Deep Neural Networks
Computational Brain & Behavior
On the importance of severely testing deep learning models of cognition
Cognitive Systems Research
Successes and critical failures of neural networks in capturing human-like speech recognition
Neural Networks
There is still little or no evidence that systematic phonics is more effective than common alternative methods of reading instruction: Response to Brooks (2023)
Review of Education