Identifying, Categorising and Recognising

We often take it for granted, don’t we? How we see something and instantly know what it is. But there’s some clever science behind it and during my MSc I conducted an experiment to demonstrate how some of it works.

Prototype theory (Rosch & Mervis, 1975) aims to explain how we recognise and identify the things around us by showing how we categorise objects, experiences and ideas.

Prototype theorists state that we hold a typical, idealised image of every concept (such as a car) in our brain (Eysenck & Keane, 2010).

When we experience something new (the latest Mini), we compare it to the prototype and if it fits the image, we categorise it as a car.  Researchers have shown that the closer the new item is to the prototype, the quicker we are to categorise it as such, and thus have been able to measure these typicality effects in terms of reaction time (Larochelle & Pineau, 1994).  For example a participant might be shown a typical item such as a Ford Mondeo, or an atypical item such as an Robin Reliant or a non-category member such as a pram; they would then have to state whether it was a car or not, and their response time would be measured.  Tests such as these have shown that reaction times have been slower for atypical items than typical ones, supporting Prototype theory.

This study aims to test the validity of Prototype theory using such a categorisation test but using a different set of typical, atypical and non-category members.

Method

This study chose to use typical and atypical countries as the category members, and cities as the non-category members:

  • 15 typical countries from list of the most visited countries (Geography.about.com, 2011)
  • 15 atypical countries from list of the least visited countries (Garfors.com, 2013)
  • 15 non-category member cities from list of most visited cities (Euromonitor International, 2011)
  • cities located in countries on either list were excluded, to avoid confusion

The lists were entered into the EPrime experiment and computerised Eprime tests were carried out.  Participants saw a + in the centre of the screen followed by a word which then had to be categorised as a category member by pressing a 1 or a non-category member by pressing a 2 on a keyboard.  Reaction times were measured:

  • 10 participants
  • All were students on the MSc Psych course at Leeds Met University
  • 8 females & 2 males
  • Aged from 22 to 57 years (mean: 36 years)
  • 45 item test for each participant

Data was exported as an Excel file and imported into SPSS for analysis.

Results

A paired-samples t-test was conducted to investigate the effects of typicality on time taken to categorise category exemplars:

  • Statistically significant difference in categorisation time for typical and atypical category members (t = 5.393, df = 9, p  < .001) with a large effect size, r  = 0.87.
  • Atypical category members took longer to categorise than typical category members

Conclusion

The country categorisation task supports the Prototype Theory

  • Prototype Theory can be used as a model to explain categorisation processes and performance.

Further implications of this study are that Prototype Theory is a valid model which can be used to study semantics and meaning change of words.

Recent categorisation studies investigate cross-cultural, goal-derived, combined category and context dependent categorisation (Davis & Love, 2010).  Future modifications of this current test could introduce the category types above to understand whether Prototype Theory can also explain the categorisation process within these areas.

However, Prototype theory has been criticised on the basis that Exemplar Theory can also explain typicality effects and can do so more effectively (Storms, De Boeck & Ruts, 2000), therefore the validity of this current test may be under question.  To understand whether which theory best explains categorisation of countries, future studies would need to run two tests, one measuring a prototype predictor variable and another measuring an exemplar predictor variable to see which one most accurately correlates with the typicality effects.

References

Davis, T., & Love, B. C. (2010). Memory for Category Information Is Idealized Through Contrast With Competing Options. Psychological Science, (2), 234. doi:10.2307/41062195

Euromonitor International. (2011). Top 100 city destinations.  Retrieved from:  http://blog.euromonitor.com/2013/01/top-100-cities-destination-ranking.html

Garfors.com. (2013). The 25 least visited countries in the world.  Retrieved from: http://www.garfors.com/2013/01/the-25-least-visited-countries-in-world.html

Larochelle, S., & Pineau, H. (1994). Determinants of Response Times in the Semantic Verification Task. Journal Of Memory And Language, 33(6), 796-823. doi:10.1006/jmla.1994.1038

Rosch, E. & Mervis, C.B., (1975) Family Resemblances: Studies in the Internal Structure of Categories.  Cognitive Psychology, Vol.7, No.4, pp. 573–605.UNWTO, 2011.

Storms, G., De Boeck, P., & Ruts, W. (2000). Prototype and exemplar-based information in natural language categories. Journal Of Memory And Language, 42(1), 51-73.

Geography.About.com. (2011). 20 most popular countries as tourist destinations. Retrieved from: http://geography.about.com/od/economic-geography/a/20-Most-Popular-Countries-As-Tourist-Destinations.htm

Leave a Reply