Friday, 22 July 2016

Open science reading list

Science has its problems, but many early career researchers (myself included) can often struggle when it comes to knowing how we can improve systems that we still very much have to operate within on a daily basis.

That said, I am a firm believer that making research readily available to others is something that we should all work towards where possible. This applies to publications, data, computer code/software and the peer review process.

The references below are taken from my own reading, but this list certainly isn't exhaustive.

All of these papers pull in the same direction. Specifically, they provide convincing evidence that open access research practices help science as well as the individual researcher.

Early career researchers, who are typically gifted very little time to get ideas off the ground and demonstrate that they have societal importance, will help their own cause by ensuring that work is readily available across multiple disciplines and beyond.

Moving forward, the next major issue for open access is no  longer whether it should be at the centre of the mainstream scholarly communication system, but how it will work effectively. 

Antelman, K. (2004). Do open-access articles have a greater research impact?. College & research libraries65(5), 372-382.

Davis, P. M. (2011). Open access, readership, citations: a randomized controlled trial of scientific journal publishing. The FASEB Journal25(7), 2129-2134.

Donovan, J. M., Watson, C. A., & Osborne, C. (2014). The open access advantage for American law reviews. Edison: Law+ Technology (JPTOS's Open Access Journal), Forthcoming.

Harnad, S., & Brody, T. (2004). Comparing the impact of open access (OA) vs. non-OA articles in the same journals. D-lib Magazine10(6).

Kousha, K., & Abdoli, M. (2010). The citation impact of Open Access agricultural research: A comparison between OA and non-OA publications.Online Information Review34(5), 772-785.

Lawrence, P. A. (2008). Lost in publication: how measurement harms science. Ethics in science and environmental politics8(1), 9-11.

PLoS Medicine Editors. (2006). The impact factor game. PLoS Med3(6), e291.

Piwowar, H. A., & Vision, T. J. (2013). Data reuse and the open data citation advantage. PeerJ1, e175.

Sandve, G. K., Nekrutenko, A., Taylor, J., & Hovig, E. (2013). Ten simple rules for reproducible computational research. PLoS Comput Biol9(10), e1003285.

Siebert, S., Machesky, L. M., & Insall, R. H. (2015). Overflow in science and its implications for trust. Elife4, e10825.

Walsh, E., Rooney, M., Appleby, L., & Wilkinson, G. (2000). Open peer review: a randomised controlled trial. The British Journal of Psychiatry,176(1), 47-51. 

Wang, X., Liu, C., Mao, W., & Fang, Z. (2015). The open access advantage considering citation, article usage and social media attention. scientometrics,103(2), 555-564.

Wicherts, J. M. (2016). Peer review quality and transparency of the peer-review process in open access and subscription journals. PloS one11(1), e0147913.

Tuesday, 5 July 2016

The effect of exam anxiety on time perception

The brief study reported below appeared in an earlier draft of my PhD, but wasn't included in the final version. The sample is small, but the results are interesting enough to merit a blog entry. If nothing else the method certainly hints towards the abundant use of smartphones in future work!


States of mind characterised by disturbed time perception (e.g. schizophrenia [Davalos et al. 2003] and some hallucinogenic states [Wittman et al. 2007]) can severely disrupt normal functioning. However, even in normal mental states, time perception is surprisingly labile. For example, during threatening events such as car accidents and robberies, many people report that events seem to pass in slow motion - as if time has slowed down. Emotions in a more general sense are also known to influence and sometimes distort time perception. Droit-Volet and Meck (2007) for example, observed that the exposure of emotional faces were overestimated compared to neutral ones.

The attentional gate model might explain why people are relatively inaccurate in making timing judgements during emotional experiences (Figure 1). In some cases, emotions demand attentional resources for coping with those events and duration estimates will be underestimated. In other cases, especially when emotions have a threatening quality, the duration will be overestimated. However, researchers have often struggled to examine this within a typical population because simulating a state of anxiety is both difficult and not ethically palatable. In this experiment, we bypassed this issue this by conducting a short experimental task outside an exam hall, where people are naturally anxious (Arana and Furlan 2016). We therefore predicted that anxiety directly before an exam would positively correlate with estimates that underestimate actual duration.

Figure 1. A pacemaker produces a series of pulses that are fed into an accumulator. The number of pulses that has been recorded for a given time span represents experienced duration. The pulse number is compared with stored representations of time periods that can be verbalized (as seconds or minutes). Attentional gate models  assume that only when attention is directed to time pulses are accumulated. 


Written consent was obtained from participants in advance of their exam, but all participants remained naive to the purpose of the study until they met with the experimenter outside the exam hall. Prior to entering an undergraduate exam hall at The University of Glasgow, participants were required to use an iPhone App (Figure 2) and adjust the rate of flashing to what they best perceived to be a rate of one flash per second (1Hz). They were also asked to give brief measures of anxiety and relief on a simple 1 to 10 scale. This procedure was repeated after they left the exam. Their exam performance was also recorded.

Figure 2. Screenshot from an iPhone app that allowed participants to adjust the rate of flashing to best match a rate of 1Hz.

13 first year psychology students studying at the University of Glasgow agreed to take part. They received one course credit in return for their participation.


Manipulation check
A paired sample t-test [t(11) = -3.05, = .011] revealed that self-reported anxiety scores were significantly higher before (= 5.67 SD = 2.10) than after the exam (M  = 2.83 SD = 2.29).  No significant changes were observed between time estimates before (M = 67.08 SD = 27.76) or after (M = 62.67 SD = 11.56) the exam [t(11) = -.521, p = .612]

Figure 3. Scatter plot illustrating a positive relationship between self-reported anxiety and adjusted rate of flashing. Red line indicates the correct response (1Hz).

A positive correlation (Figure 3) was observed between self-reported levels of anxiety and participants adjusted rate of flashing [r(11) = .82, p < .01].  A second correlation coefficients showed no significant relationship between anxiety or rate of adjusted flashing post-exam [r = -.15, ns].


In the first instance, this experiment appears to support previous research that increased levels of anxiety show a strong correlation with underestimating duration. In other words, time appeared to speed up, but only for participants who rated their anxiety above 5 (out of 10). The reverse was true for those who felt relatively relaxed. As expected, higher levels of anxiety associated with a forthcoming exam appear to require attentional resources and the duration of flashes was therefore underestimated (Droit-Volet and Gil 2009). What remains unclear however, is whether participants estimates were guided by the flashes themselves or the gap between each flash. This effect could be partly driven by people who are more anxious in general, but this seems unlikely given that the effect disappeared post-exam where participants were no longer anxious. However, it is not possible to rule out practice effects completely. 

Cognitive-processing mechanisms appear to guide our experience of duration. While the perception of time is strongly linked to our subjective-well being, it also varies considerably depending on specific emotional states (Wittmann and Paulus 2007). The feeling that time passes quickly or slowly can be a fair indicator of both psychological distress resulting from an inability to focus meaningful thoughts, but it may also help focus our attention on important activities in other contexts that require considerable cognitive resources. 


Arana, F. G. and Furlan, L. (2016). Groups of perfectionists, test anxiety, and pre-exam coping in Argentine students. Personality and Individual Differences90, 169-173.

Davalos, D. B., Kisley, M. A. and Ross, R. G. (2003). Effects of interval duration on 
temporal processing in schizophrenia. Brain and Cognition. 52, 295–301.

Droit-Volet, S. and Meck, W. H. (2007). How emotions colour our perception of time. Trends in Cognitive Sciences11(12), 504-513.

Droit-Volet, S. and Gil, S. (2009). The time-emotion paradox. Philos Trans R Soc Lond B Bio Sci. 364, 1943-1953.

Loewenstein, G. F. (1996). "Out of control: Visceral influences on behavior. Organizational Behavior and Human Decision Processes. 65(2), 272-292.

Wittmann, M. and Paulus, M. P. (2007). Decision making, impulsivity and time perception. Trends in Cognitive Sciences. 12(1), 7-12.