Ichijo, E., Lee, K. S., Stantić, M., De Castro, I., Murphy, J., Vafeiadou, A., Banissy, M. J., Catmur, C., & Bird, G. (2025).
Good studies in psychology are
generalisabile and replicable.
That means that:
– the group of participants really should represent the general population – generalizability.
– and the results should be able to be done again with very similar results – replicability.
But, results show that people who sign up for psychological studies typically don’t represent the general population, usually due to age and background. This can be taken into account with some good statistics, but low-level psychiatric symptoms are not usually checked for. This research follows a series of studies to explore why it is important to do so.
Study 1: Over 850 people, with no psychiatric diagnosis or medication, were recruited and completed online surveys of possible symptoms. These results were compared to national data on the general population. They found that these people joining studies had significantly higher rates of various psychiatric symptoms than the general population. But, this was done during the COVID-19 pandemic, and that needed to be accounted for.
Study 2: Study 2 replicated Study 1, but this time only using data from before the pandemic. Results remained the same: those people signing up for studies had significantly higher rates of symptoms compared to the general population.
Study 3: The final study asked a new group of participants to complete the psychiatric surveys as well as some cognitive ability tests. Participants with no symptoms had the most consistent reaction times but were slower to remember things and didn’t do as well on cognitive tests. Participants with some symptoms were the fastest at remembering things and did the best on cognitive tests. These results show that you can group people’s performance by significant psychiatric symptoms, and that these psychiatric symptoms can impact cognitive ability.
The researchers explain that just because someone doesn’t have a formal psychiatric diagnosis, it does not mean they don’t have some of these experiences. This means that when scientists compare people with and without formal diagnosis, the ‘not-diagnosed’ group will most likely include people who still have some indicators. It may be wise to check for these so that statistics can be used to get a clearer result to ensure both replicability and generalisability.
summary by Hannah