language and reading

Question 1

What is a word?

Answer 1

A form with a function

Question 2

What is the difference between spoken word form and written word form?

Answer 2

Spoken- sequence of phonemes organised into syllables, some have stress patterns.
Written- a sequence of symbols (graphemes) made of lines, curves or strokes

Question 3

What is lexical access?

Answer 3

Components of reading:
Identify letters and represent their sequence -> identify words -> retrieve (syntactic class and word meaning) -> interpret sentence structure -> interpret sentence meaning -> interpret intention of speaker/writer

Question 4

What is the relation between spelling, pronunciation and meaning?

Answer 4

Relation between form and meaning is arbitrary but translation from spelling to pronunciation varies across language- it is semi regular for English

Question 5

What are the different levels of analysis in word identification?

Answer 5

Levels of analysis: experiential, computational or functional (representations, processes, modules and architecture). Neural- how neurons implement computations and where they are.
We can manipulate certain factors that influence the speed or ease that we can read or understand.

Question 6

What are the two kinds of behavioural measure?

Answer 6

(artificial) lab tasks designed to exercise and capture a component process: typically discrete stimulus -> response tasks. Allowing for accuracy and/or reaction time to be measured to each stimulus.
On-line measures made during continuous natural performance of the skill

Question 7

What is the word superiority effect?

Answer 7

greater accuracy of letter idenficication in the context of a word than of a non matched word

Question 8

What are frequency effects?

Answer 8

RTs for lexical decision, semantic categorisation and naming shorter for words that are more frequent in the language

Question 9

What are sentence context effects?

Answer 9

RT’s for lexical decision and naming are shorter when a word is presented in a sentence context of which it is a plausible continuation

Question 10

What did Treisman and Gelade measure in relation to the visual search task?

Answer 10

They measured the time taken to search for a target object that differed from the background objects and varied the number of objects in the display.

Question 11

What is a task set?

Answer 11

An appropriate organisation of processes to carry out a particular task

Question 12

How can executive or control processes be measured?

Answer 12

Get ppts to switch back and forth between tasks and measure the effects of this on performance

Question 13

What are two main functions of selective attention?

Answer 13

Defensive filtering- protecting higher level limited capacity systems from overload (Broadbent).
Positive selection for action- prioritising one of several possible objects or sources for action or further processing. Feature of integration or ‘binding’ combining the properties of an object anaylsed in different cortical maps. Evidence: visual search for conjunction vs single feature targets harder. Illusory conjunctions.

Question 14

What is the feature integration theory?

Answer 14

Regions of visual cortex are specialised for the local analysis of different attributes such as colour and form. Feature searches can be done pre attentively for targets defined by only one featured. Focused attention is needed to bind features of the same object and only then do we perceive it as an object.

Question 15

What does the emotional Stroop tests measure?

Answer 15

Emotional stroop uses an indirect measure of emotional bias in psychological disorders as opposed to recognising emotions in faces for example (a direct measure)

Question 16

What does the emotional Stroop test tend to find?

Answer 16

Time taken to name colour increases as attention to emotion word increases
Patients with depression tend to be slower to name the colour of negative words (indirect negative bias)

Question 17

What does the emotional Stroop test show in disorder relevant words?

Answer 17

Threat words in anxiety, injury related in PTSD, exam related in students.
Both relatedness to current concern and negativity cause greater interference in patients

Question 18

What are some limitations of the emotional Stroop test/ alternative explanations?

Answer 18

Unclear what it really measures however not a very clean measure- may be other explanations e.g. emotional reaction triggered by reading certain words. Can interfere with ruminaitons etc. however, some studies have reported effects with subliminal presentation of words. Suggesting it may be a more automatic process of attention allocation rather than people deliberately trying not to look at a word.

Question 19

What is the mean fixation duration?

Answer 19

200ms

Question 20

What is the mean saccade length?

Answer 20

8.5 characters

Question 21

What is the relationship between words, letters and context?

Answer 21

Higher frequency words are recognised more easily and faster- letters are more easily recognised in the context of a word and words more easily recognised in the context of a sentence

Question 22

What are the two theories for why frequent words are recognised more quickly?

Answer 22

Foster’s serial search model- where we search the lexicon in order of frequency
Style parallel matching process- can also account for frequency where most used detectors are the most ;sensitive’

Question 23

What is the serial search model?

Answer 23

Encode a spelling pattern
Compare one at a time to each word-form stored in the mental dictionary
If a match is found, we retrieve meaning and/or pronunciation, if not, we continue the search.
For efficiency, the search is in order of frequency in the language, hence a match will be found faster for a higher frequency word.
This does however require many thousands of successive comparisons per second- too fast for neurons but can be done by a computer.
Mental lexicon of word forms is divided into ‘bins’. Serial search within that bin (frequency ordered).

Question 24

What is the interactive activation (parallel matching word detector model)? Mclelland et al

Answer 24

Model consists of 3 layers- first layer detects the features, feeds the info forwards, depending on which features are fed forwards, the model determines which letter it is, these letter activations get fed forwards to the third layer and depending on activation, the word is found. Only at the end does the word get identified. Some top down activation of letter units can explain the effect of context.

Question 25

What frequency of words produce a homophone effect?

Answer 25

Low frequency e.g. if word HAIR and HARE ar high frequency, homophone effect tends to disappear

Question 26

Summary q: how do we identify words (decide that a letter string is a word)?

Answer 26

• By serially going through the lexicon (words are ordered in bins according to frequency)
• parallel search (first identify features, then letters, then words)

Question 27

How do we go from print to meaning?

Answer 27

Via pronunciation or not

Question 28

What sentence structure clues help with word comprehension?

Answer 28

Word order- e.g. the wife admired the man vs the man admired the wife
Function words (as opposed to content words)- small fixed set of grammatical words that do structure signalling jobs e.g. the/a/an are ‘determiners’ whereas who/which/that are relative pronouns.
Word modifying morphological inflections (not v common in English).
Syntatic structure provides one kind of clue to the meaning-some syntatic structures without much meaning means you won’t be able to understand even though individual words have meaning.

Question 29

What ambiguity do words with several distinct meanings have?

Answer 29

Lexical ambiguity

Question 30

What ambiguity do words with variations of grammar have?

Answer 30

Syntactic ambiguity

Question 31

What is ambiguity of reference?

Answer 31

Doubt over who/what the sentence is referring to e.g. Paul gave John his hat

Question 32

What three steps do we need to take to comprehend the text?

Answer 32

1. Interpret the speaker/writer’s intentions
2. Fill the gaps with inferences
3. Extract the meaning from the text as a whole

Question 33

What is the difference between direct speech acts, indirect speech acts and irony/metaphors?

Answer 33

Direct- intention is revealed by the literal meaning of words
Indirect- literal meaning is not the intended one, not usually a request for info
Metaphor- literal meaning is not the intended one

Question 34

How do we infer the intended meaning of words?

Answer 34

We have to infer the unspoken based on: the extra linguistic context (immediate environment, assumed status, intentions), body language/gestures/expressions, Linguistic context (what has been said before), general knowledge, grammar, capitalization, properties of things e.g. personal charctetrisics.

Question 35

What evidence did Graham (1979) find that at least some inferencing is automatic?

Answer 35

tested cued recall for lists of sentences. We infer automatically and remember what we infer as if it were explicitly stated (even when instructed to remember word for word).

Question 36

What is the difference between lexical and syntactic ambiguity?

Answer 36

Lexical- the presence of two or more possible meanings within a single word e.g. ‘I saw her duck’- e.g in sport or her actual pet duck
Syntactic- the presence of two or more possible meanings within a single sentence of sequence of words e.g. ‘the chicken is ready to eat- - chicken as in cooked meat or actual live chicken

Question 37

What 3 ways can we solve ambiguities?

Answer 37

Minimal commitment strategy- Postpone interpretation until all potentially disambiguating, but this is not plausible as priming of word recognition by incomplete sentence contexts indicate comprehension is incremental
Serial strategy- construct the most probable interpretation, backtrack if it turns out to be the wrong one
Parallel strategy- construct multiple interpretations in parallel, delete those that don’t work out

Question 38

What do all 3 strategies that try to solve ambiguities require?

Answer 38

They require working memory to hold the input, the words and their order or represent the output

Question 39

How do we process local lexical ambiguities?

Answer 39

‘Semantic priming’ as an index of activation of meaning e.g. primed with bread- probe is butter, as this provides a measure of activation of the meaning of the prime.

Question 40

How does the location of disambiguating context change the processing cost?

Answer 40

If no disambiguating context BEFORE and meanings are of EQUAL frequency then there are longer fixations on ambiguous word.
If there IS disambiguating context before the ambiguous word then fixations on ambiguous word are no longer than for unambiguous so suggests both meanings are not activated.

Question 41

What is the current conclusion on lexical ambiguity resolution?

Answer 41

meanings of ambiguous words activated in parallel but not with equal strength. The relative strength of an action depends on the degree of contextual constraint available and the frequency of use.