Language & Thought
Remember, use this page at your own risk! Follow up the theories and models mentioned to build up your own reading, but, having read these, come to your own conclusions about them. These evaluations may be referenced, but they may not perfectly reflect the study findings.
Although language and thought has a long history in philosophy, the most well-known consideration of this within psychology is the Sapir-Whorf Hypothesis; this proposes that our thoughts and perception are moulded by our language. From this, several versions of the hypothesis have been created.
The Sapir-Whorf Hypotheses
According to Regier & Kay (2009), there are two versions of the Sapir-Whorf Hypothesis:
Relativist stance or the Strong version - this proposes that language determines our thoughts. If you can't express it in language it cannot be thought about. This means that learning another language fundamentally changes your thought processes. This theory is almost entirely dead now, having not been founded on research evidence. the theory would predict, for example, that cultures who have only 2 words for colour, such as dark and light (there are cultures like this) would only see the world in 2 colours. Also, it is clear Bilinguals do not think in a fundamentally different way to monolinguals.
Universalist stance or weak version - This proposes that language influences thought through our perception. How we perceive the world is modified slightly by the language we use to describe it. For example, this theory would argue that Eskimos have several words for snow, so their perception is affected insofar that they are better at distinguishing between types of snow. This stance is still an active research area.
The "strong version" of the Sapir-Whorf Hypothesis is known as linguistic determinism; whereas, the "weak version" is known as linguistic relativity.
These hypotheses have been evaluated researching several topics, such as colour and space perception, memory, emotion, and so forth.
Does language influence perception?
Carmichael, Hogan & Walter (1932) found that How an object is described affects the way it is recalled. they gave participants ambiguous drawings and one of 2 descriptions, and then later asked the participants to re-draw the image. Participants drew significantly different drawings based on how the object was described. for example, one of the items was a curved line that looked like the letter C, but could also just as easily be a crescent moon. if told before that it was the letter C, participants 'remember' it being much more like the letter than it actually was.
Colour perception has been one of the most active areas of lingistic relativity research, as its one of the easiest to investigate. In general colour supports the weak version of the hypothesis, as colour is a matter of perception, not processing.
Research by Regier & Kay (2009) has found that language may shape colour perception in the right visual field via activation of language regions in the left hemisphere. Furthermore, linguistic categorical perception has also been located to the left hemisphere. Correlational evidence for this can be found in pre-linguistic infants. Based on colour-categorical perception tests, they showed little to no activity in their right visual field; rather, activity occurred in their left visual field. This suggests that language may have an impact on perception insofar that the activity switches over to the left hemisphere, and therefore to the right visual field. In other words, the categories that were used pre-linguistically occurred in the right hemisphere, but then linguistic categories may replace these, and thus generate more left hemisphere involvement in colour-categorical perception.
Thierry et al (2009) also found that certain areas in the left hemisphere have top-down influences on colour perception. This also implies that language shapes perception via the new categories learnt to help distinguish colours in environment.
However, a study by Roberson et al. (2000) found that does not appear to be a universal basis for colour perception, which suggests that some categories within language, at least, may be exclusive to that language.
Lupyan (2012) proposed the label-feedback hypothesis. In short, this argues that the labels we use are part of a feedback system in our perception, so it can affect our perception both profoundly and transiently, but in separate ways.
He observed that the language-thought debate is complex because observations find that language can have deep, long-term effects on our perception, but it can also be subject to superficial change. For example, trends for use of words (e.g. 'groovy' in the 1970s) come and go, and technology may have only served to speed up such a process. But then research on colour perception (Thierry et al., 2009) shows that how we label colours changes our perception of them, in particular, our sensitivity to them. How can such a contradiction exist? Lupyan believes the label-feedback hypothesis may help provide a more interactive explanation for the relationship between language and thought.
Bleach effects - these refer to the verbal interference that can eliminate cross-linguistic perceptual differences. This is an example of language being easily prone to change.
The label-feedback hypothesis argues that language can affect perception and categorisation by having an influence on their continual processes. These processes are so quick and automatic that they come to us as thoughts. Therefore modulations on these means a modulation on our thoughts. But does this necessarily mean that language moulds thoughts to the extent that all of our behaviour is affected by it?
Previous research by Goldrick & Hendrickson (2010) found that language does indeed alter perception. In particular, they found that objects that lay near category boundaries (just about fulfil the criteria for that category) could be perceived as being more similar or different depending on how it is labelled. Their 'perceptual space' is warped, and therefore our sensitivity to the similarities and differences of these objects changes.
This is also supported by cortical plasticity (James, 1890; Recanzone, 2000) in the visual field, or Wernicke's area. Our sensitivity to stimuli (in this case, visual) is altered by reorganisation of our cortical representation for it. More representation means we are more sensitive to an objects' differences, and vice versa.
This is further (potentially) supported by a hypothesis for object recognition, called Rapid Visual Categorisation (Fabre-Thorpe et al, ????). They argue that object recognition and creating retinotopic maps for these objects undergoes a two-stage process:
- Categorisation first - Details later
This is in line with the top-down processing model of the Label-Feedback hypothesis.
By what mechanisms would language affect thought?
One idea is based on the idea that we do not think in grammatical language, but in a sort of mentalese. However when we express a thought we fit it into the constraints of our language, and through repetition (phonological encoding) we might come to remember what we said as opposed to the original thought.
It may be that language influences what we pay attention to, and that influences our perceptions. We tend to pay more attention to thing we have labels for. When we are given a label for something we activate concepts for it. These then interact with sensory stimuli to produce the perception.
Criticisms and Controversies
This may or may not be specifically useful for the 'language' question in the exam, but it might be useful for making links between the topics and therefore improving and economising the information you have learnt for the whole module.
How can we make use of other theories and models learnt in this module and apply them to language?
- ACT* Model
- Working Memory Model
- TLC Theory
- Attentional processes: Procedural and declarative memory, 'choking under pressure', Norman's expertise criteria, Anderson's skill acquisition
- PSY249 - Danielle Matthews' lectures