Faculty and Staff
We value our faculty and staff members and are always looking for talented people to join our thriving team. Excellent benefits, diverse career opportunities, and a true community spirit are just some of the reasons you should consider joining our team of talented, dynamic faculty and staff. For a full list of current openings, visit UBC’s Staff & Faculty Careers page.
Each academic year, we have openings for Graduate Teaching Assistants (GTAs) and possibly Undergraduate Teaching Assistants (UTAs) for Economics, Philosophy and Political Science courses in the September and January terms.
Teaching Assistants may be involved in the following:
- marking midterms, assignments, tests, quizzes and exams
- leading discussion periods and tutorials
- invigilating tests and examinations
- providing academic assistance to students during office hours
The hours will not exceed an average of 12 hours per week. Wages, as stated in the current BCGEU Collective Agreement, are as follows:
- Graduate TA (PhD program): $34.72 per hour ($13,331.20 annual rate September – April)
- Graduate TA (Master’s program): $33.44 per hour ($12,842.64 annual rate September – April)
- Undergraduate TA (Bachelors program): $17.82 per hour ($6,844.72.16 annual rate September – April)
- Markers: $17.32 per hour
Subject to funding, the Department of Economics, Philosophy and Political Science has part-time TA positions available in 2021S and 2021W Terms. Qualifications preferred include an understanding of the discipline and good grades. UTA’s must be a major in the discipline, have fourth-year standing, and must maintain a minimum of 18 credits in the Winter session (2020W) and a minimum of 9 credits in the Summer session (2021S). Part-time students can also be considered.
2021/2022 GRADUATE TEACHING ASSISTANT (GTA), UNDERGRADUATE (UTA), and Markers
DEADLINE: April 30, 2021. Late applications will not be considered.
Please contact firstname.lastname@example.org for inquiries.
UBC Okanagan hires on the basis of merit and is strongly committed to diversity within its community and especially welcomes applications from women, visible minority group members, Aboriginal persons, persons with disabilities, persons of any sexual orientation or gender identity, and others who may contribute to the further diversification of ideas. However, Canadians and permanent residents of Canada will be given priority.
Paid research assistant opportunity (1 or 2 year position)
Project: Perceptual Bias and its Role in Racial and Gender Discrimination
Principal Investigator: Madeleine Ransom, Assistant Professor of Philosophy, Department of Economics, Philosophy, and Political Science at UBCO
Duration: September 2021 to August 2022. The position is renewable for a second year (September 2022 to August 2023) though this is not a requirement. Students expecting to graduate soon after August 2022 are eligible to apply.
Remuneration: $19,952 per year, plus all expenses paid to two conferences (likely the SPP in year one, Pacific APA in year two). Note that holding this RA position comes with the expectation that the stipend replace TA work for the year. If the student wishes to take on additional TA work then this should be discussed prior with the PI in order to ensure that it does not interfere with this project.
Role of research assistant:
- Read two articles a week
- Meet once a week for an hour via zoom to critically discuss weekly readings and how they fit into the larger research topic
- Provide brief (no more than one page) summaries of readings
- Compile bibliography in Mendeley (support and tutorials available through UBC library)
- Spring 2022: assist organizing a small online workshop at the Brains Blog e.g. Manage participant list, answer email questions, remind participants of schedule
- June 2022: Attend the Society for Philosophy and Psychology annual meeting (all expenses paid), with the option to attend the SPP one-day workshop, usually on a current hot topic of research. Optional: the student may wish to submit a paper to the conference, so they have an opportunity to present their work if it’s accepted.
Who may apply: any graduate student in philosophy at UBC, though preference will be accorded to PhD students. The person must plan to remain a graduate student until at least April 2022.
Application process: email your academic CV (including a list of all graduate courses taken) and a short cover letter explaining your interest in the project and how it fits with your own research interests and/or work. Applications should be sent to email@example.com
Application deadline: end-of-day August 20th, 2021.
Bias impacts who gets hired, approved for a bank loan, convicted of a crime, granted parole, and even –in extreme cases such as police shootings and access to quality medical care– who lives and who dies. Yet the phenomenon of bias is not well understood. While many varieties of cognitive and algorithmic biases have been identified, little work has been done to develop overarching taxonomic categories that can help further academic understanding of the complex interdependencies between kinds of biases. For example, perceptual bias –commonly operationalized in perceptual psychology as the average size of the difference between perceived and true perceptual stimulus values– itself remains under-theorized despite its potential contribution to racial and gender bias. The main objective of my project is to develop a comprehensive conceptual framework for perceptual bias and –through two case studies– articulate its contribution to discrimination.
My first case study – and the topic of my first article – examines the empirical phenomenon of categorical perception. In categorical perception, it is easier to distinguish between two members of different (learned) categories than it is of two members of the same category, even though the magnitude of the physical differences across all objects has been controlled for (Harnad 1987, 2003). This is thought to be because objects in the same category actually appear more visually similar, and objects in different categories appear less similar (Goldstone and Hendrickson 2010). A paradigmatic example of this is colour perception. Despite the fact that a rainbow contains a continuous spectrum of colours, we perceive it as being composed of relatively discrete categories such as ‘green’ and ‘blue’, and have more trouble distinguishing shades of blue from each other than from shades of green (Bornstein and Korda 1984; Pilling et al. 2003; Roberson and Davidoff 2000; Roberson, Pak, and Hanley 2008).
In cases like the example of colour, our perceptual experience does not accurately track the world in that it is biased with respect to low-level sensory input. Nevertheless, there is a sense in which it is rationally unbiased, or epistemically justified, in that parsing the world into discrete categories makes the computationally intractable tractable, and that categorization allows for the accurate tracking of categorical properties such as ‘blue’ rather than simply representing a spectrum of colours. However, categorical perception effects have also been found with respect to race and gender – just like we perceive the difference between blue and green as more pronounced than it really is, so to do we experience the perceptual differences between our learned race and gender categories as larger than they are (Levin and Angelone 2002; Campanella, Chrysochoos, and Bruyer 2001). Here we might question whether our categorizations continue to be epistemically justified or rational. The answer is complicated by philosophical debate over whether race is a biologically real category (Sullivan and Tuana 2007; Zack 2014; Msimang 2020; Andreasen 2000; Glasgo 2003; Maglo 2011; Gannett 2010; Pigliucci and Kaplan 2003; Q. Spencer 2018, 2014). If it is not, then ‘racial’ categorical perception is not actually tracking race, but something different, and an expanded analysis of perceptual bias allows for a more precise diagnosis of how and where the perceptual process goes wrong.
Even if we were to conclude that such categorization does track the real property of race (for example, it might be socially constructed rather than biologically real), there is also the question of the morality of such categorization: though the perceptual categorization of colours is not morally problematic, the categorization of people may be. While racial and gender perceptual categories can of course be associated with harmful and derogatory beliefs, the mere fact of categorization may promote the treatment of people as objects, diminishing the possibility of engaging with people as individuals (Basu 2019b, 2019a; Munton 2019). Moreover, if we adopt the view that race or gender are socially constructed categories – ones that have led to and undergirded systems of racial and gender oppression (Haslanger 2000, 1995) – then our continued perceptual categorization may render us complicit in perpetuating these systems. Here again there is some literature in machine learning on what are called ‘looping effects’ that is useful in precisifying the variety of bias at work here (Barocas, Hardt, and Narayanan 2018).
My second case study, and topic of my second article – disfluency bias – occurs when people at the margins of a given perceptual category are subsequently judged or treated more negatively as a result of the relative difficultly in perceptually categorizing them. The fluency with which an object is processed is altered in part by how prototypical an object is. Prototypes are stored representations of the central tendencies of categories that have been hypothesized to allow for perceptual categorization (Rosch 1973; Goldstone and Barsalou 1998; Barsalou 1999; Hampton 2006). The more prototypical an object is, the greater the fluency with which it is processed in perception, and this results in the generation of positive affect (or negative if the processing is disfluent), which then gets bound or associated with the object of perceptual processing (Bullot and Reber 2013; Reber, Schwarz, and Winkielman 2004; Winkielman et al. 2003). The result is that, in many cases, an evaluative property is attributed to the object, or the object is judged more positively or negatively.
The most well studied disfluency bias is the halo effect, where a person’s (lack of) physical attractiveness or beauty influences subsequent judgments of character traits such as kindness, honesty and intelligence (Nisbett and Wilson 1977; Keeley et al. 2013; Feeley 2002; Murphy, Jako, and Anhalt 1993; Eagly et al. 1991; Ramsey et al. 2004). This leads to a form of morally problematic bias called ‘lookism’ that has real psychological and economic consequences, including decreased parental attention, lower salaries and fewer job opportunities (Minerva 2017). There is also emerging research regarding gender non-conforming individuals: people who are not easily categorized as male or female (Lick and Johnson 2013; K. L. Johnson, Lick, and Carpinella 2015; Lick and Johnson 2013, 2015). Similar to the case of lookism, subjects produced more negative evaluations for gender non-conforming individuals. My case study will focus on the perceptual foundations and ethical implications of this emerging research.
One aspect of this phenomenon of particular interest is that, because gender nonconforming individuals make up a minority of the population, then our prototypes of the perceptual categories of ‘male’ and ‘female’ may actually reflect the true population average. In such cases our categories are not biased according to a statistical standard. Nevertheless, they are arguably biased according to a moral standard, given the negative consequences of disfluent processing. Here I will connect my analysis of the phenomenon to the rich philosophical discussion from the implicit bias literature on whether or not we can be held morally responsible or blameworthy for our biases, given that they are not under our direct control and that we are often not aware of their influence on our beliefs (Isaacs 1997; Harman 2011; Graham 2014; Mason 2015; Brownstein & Saul 2016b).
In my third article I will provide a systematic and comprehensive framework for perceptual bias by drawing upon emerging taxonomies in computer science for thinking about algorithmic bias, as touched on above (Danks and John London 2017; Glymour and Herington 2019). While these taxonomies do not map on perfectly to human perceptual bias, they nevertheless provide a useful point of departure because machine learning involves many of the same inputs and processes as those involved in perceptual learning, given that both are varieties of statistical learning. I expect that working through my two case studies will inform the analysis, both by grounding it in concrete examples and by deepening my thinking on the matter as I work through the examples.