Judy Slee Seminar Series

Date & time

4–5pm 3 July 2018


Peter Baume Building 42A Room 2.01

Tuesday 3 July 2018 This Seminar will comprise of four 10 minute presentations, followed by 5 minutes question times.

Presentation 1 & 2 by Sheri Kim.

Sheri is a third year PhD candidate at the Research School of Psychology, the Australian National University. She completed her undergraduate degree in Psychology in 2013 with Class I Honours and University Medal. Her PhD research is about research synthesis. She is especially interested in effect sizes, and applications of research synthesis in health research. She is supervised by Professor Michael Smithson, Associate Professor Bruce Christensen, Dr Dave Pasalich.

1 Title: The ins and outs of effect sizes: Effect size classification

2 Title: Bayesian Meta-analysis of Studies using Cohen's d in R.


Presentation 3 by Timothy Hatfield.  Timothy is a PhD. candidate at the Research School of Psychology at The Australian National University. He graduated with honours from the University of New England with a Bachelor of Psychology. His primary research interest is the psychology of physical symptoms, with a particular emphasis on interoception and affective disorders.

3 Title: Autism and interoception: A fragmented internal world


Presentation 4 by Toby Keene. Toby  is an Intensive Care Paramedic and health service manager. He joined the ACT Ambulance Service in 2001, qualifying in 2004. Since then, he has gained experience as a flight paramedic, clinical educator and operations manager. In 2007, he completed a 12-month attachment to the Australian Federal Police as an operations support paramedic. He has deployed on numerous disaster relief operations in Australia and overseas. Toby’s tertiary qualifications include a Graduate Certificate in Aeromedical Retrieval, Master of Public Health, and is a PhD candidate at the Australian National University.

4 Title:  Fuzzy trace theory predicts paramedic diagnostic decision better than fast and frugal heuristics in simulated patients. Toby Keene* and Kristen Pammer**, *Research School of Psychology, Australian National University, ** School of Psychology, the University of Newcastle



Presentation 1: The ins and outs of effect sizes: Effect size classification.

The Replication Crisis has revived discussion on the need to improve the research methods used in psychological research. Along with this discussion, there has been a re-emphasis on the benefits of using effect sizes. However, the use of effect sizes appear to be a “checking the box” process, with researchers simply opting for well-known effect sizes such as Cohen’s d and eta-squared (Fritz, Morris & Richler, 2012) and a general lack of interpretation alongside the effect size reported (Sun, Pang & Wang, 2010). Researchers’ theoretical understanding of effect sizes remain underdeveloped (Kelley & Preacher, 2012). Incorrect use of effect sizes may result in inaccurate or confusing information (e.g., McGrath & Meyer, 2006; Smithson & Shou, 2016), so researchers need clear guidance on which effect size to use. One way that methodology researchers attempt to gain and communicate clarity about effect size indices (ESIs) is to classify the indices into groups. However, classification systems are not exhaustive (they do not include all ESIs), the categorisations often conflict, and classification systems do not fully inform the researcher about the ESIs (they ignore some characteristics of effect sizes that may be important to consider). This presentation explores how the statistical/mechanical and functional aspects of effect sizes can help us understand the conflicting nature of effect size classification. This presentation  suggests a unified framework utilising these aspects of effect sizes and their classifications to guide researchers through the zoo of ESIs to choose, report and interpret appropriate effect sizes for their research.


Presentation 2:  Bayesian Meta-analysis of Studies using Cohen's d in R.

Bayesian meta-analysis has several key advantages over frequentist meta-analysis. First, a Bayesian framework theoretically utilises the correct conditional probability, and practically allows evidence for the null hypothesis. Second, the posterior distribution and credible intervals are intuitively interpretable. Third, data can be added as new participants or studies appear, which is particularly important in living meta-analyses (Elliott, et al., 2017; Simmonds, Salanti, McKenzie & Elliott, 2017). There already exist examples of how to apply a Bayesian framework to meta-analysis (e.g., Scheibehenne, Jamil & Wagenmakers, 2016; Smith, Spiegelhalter & Thomas, 1995; Sutton & Abrams, 2001). However, these examples only utilise Bayes factors and odds ratios or risk differences. There are no studies that demonstrate how to apply Bayesian meta-analysis on commonly used effect sizes in psychology such as Cohen’s d. This presentation  demonstrates a Bayesian random-effects meta-analysis of studies that use Cohen’s d. The meta-analysis results in an overall effect size and its credible interval, and an estimate of the between-studies variance. The analysis is conducted using Stan in R.


Presentation 3:  Autism and interoception: A fragmented internal world
Autism Spectrum Disorder (ASD) is associated with atypical functioning across multiple sensory modalities. Although difficulties in the perception of internal bodily sensations (i.e. interoception) can be inferred from the first clinical descriptions of ASD (Kanner, 1943), research to date has primarily focused on external sensory processing. In fact, only in the past 6 years have researchers sought to empirically examine the way in which people with ASD attend to and process internally-derived signals. While the contemporary studies have made some progress in examining the subjective, objective, and metacognitive aspects of interoception in ASD, the relationship between ASD status and interoceptive dysfunction is still unclear. Furthermore, the results of these studies were evaluated without consideration of the available neuropsychological frameworks for the perceptual processing of people ASD. In this presentation, we suggest that the Weak Central Coherence (WCC) framework (Frith, 1989) may offer useful insights into the manner in which interoception operates at the perceptual level in people with ASD. In particular, this conceptualisation of interoception in ASD posits that people with the disorder may demonstrate a reduced capacity to integrate local interoceptive information into coherent global feeling states (Hatfield, Brown, Giummarra, & Lenggenhager, 2017). Crucially, a potential corollary of this impairment is uncertainty in the interpretation of emotions and other feeling states in self and others. While the established methodological approaches to measuring interoception might frustrate the implementation of such a local–global approach, we suggest that data from novel topographical approaches to sensation reporting might overcome the challenges.


Presentation 4: Fuzzy trace theory predicts paramedic diagnostic decision better than fast and frugal heuristics in simulated patients.

Australian paramedics (n=129) and paramedicine students (n=127) participated in two experiments assessing fuzzy trace theory (FTT) and fast-and-frugal heuristics (FFH) in clinical decision making. Experiment 1 exposed participants to written clinical vignettes, designed to elicit intuitive decisions. Experiment 2 exposed participants to two-part written vignettes. Participants recorded impression and final diagnosis, and rated confidence and perceived typicality.

Objective likelihood partially predicted impression but not final diagnosis with no effect of experience. Perceived typicality predicted neither impression nor diagnosis. Framing effects were observed in experienced but not less experienced paramedics.

These studies provide additional evidence for FTT in an applied setting.

for each speaker.

Updated:  22 July 2018/Responsible Officer:  Director, RSP/Page Contact:  Web Admin, RSP