Hugfræði - lokapróf

What was the aim of introspection, as used by early psychology researchers in late 19th century Germany?

• it was used to verify that the tools the researchers used to measure behaviour were properly calibrated

• It was used to assess the validity of the theory that conscious experience was primarily visual, with only limited contributions from auditory and other senses

• It was used an in effort to better understand the way the mind functioned, for example by examining how different stimuli produce different conscious experiences

Saul Sternberg (1966) studied people's ability to determine whether a single number had been part of a small set of numbers they had been asked to memorise. What relationship did Sternberg find between the size of the set (i.e., how many numbers it contained) and the time people took to complete the task?

• There was no relationship: People's responses seemed to be similar irrespective of the size of the set

• There was a logarithmic relationship: People's responses took longer as the size of the set increased, but the effect "levelled off" for lager sets.

• There was a consistent linear relationship: As the size of the set increased, people's responses took longer

What is one way in which localist and distributed representations of information differ?

• In localist representations, the information being accessed can be inferred from the activation of single cells; in distributed representations it cannot

• Localist coding can only store a limited number of representations, whereas distributed coding allows for the storage of an unlimited number

• Localist representations can be formed through changes in the strength of connections between neurons, whereas distributed representations require the recruitment of entirely new cells

Some researchers use event-related potentials (ERPs), collected via electroencephalography (EEG), to inform their study of cognitive processes. What is one of the major advantages of EEG as a research tool in this context?

• It provides unambiguous information about the brain structures involved in the relevant neural event

• It allows for very precise measurement of the time at which a neural event occurred.

• It offers an insight into the way in which different parts of the brain are connected to one another

What is a major property of connectionist of theories of cognition?

• They use the properties (e.g., strength, number) of connections between neuron-like elements as their foundation.

• They are focused on understanding how different regions of the brain are connected to one another

• They attempt to explain how social relationships and connections affect the way people think in certain situations

Which of the following is an example of stimulus-driven attention (i.e, exogenous control)?

• Choosing to focus on a sound that is coming from your left, while ignoring sounds coming from your right

• Searching for a friend in a crowd, by looking for the color of the top you know your friend is wearing.

• Being distracted by flashing advertisement as you walk down a busy shopping street

What is a major difference between Treisman's attenuation theory and Broadbent's filter theory?

• According to Treisman's theory, we take in similar information irrespective of what we are attending to, but our ability to act on unattended information is attenuated (i.e., reduced)

• According to Treisman's theory, we use attention to attenuate the signals from stimuli that would otherwise be unpleasantly intense (e.g. loud noises, bright lights)

• According to Treisman's theory, attention to different channels is not all-or-none, but graded (e.g., we can deploy 80% of our attention to one stimulus, ant the other 20% to another).

In the Posner cueing task, people are given a cue indicating that a stimulus will appear to the left or the right of fixation. When the cue is valid, the stimulus subsequently appears in that location; and when it is invalid, the stimulus appears on the opposite side. Finally, a neutral cue provides no information about where the stimulus is likely to appear. What findings are typical in such a task?

• People are faster to respond to invalidly-cued stimuli than to neutrally- cued stimuli

• People are faster to respond to validly-cued stimuli than to neutrally-cued stimuli

• People are slower to respond to validly-cued stimuli than to invalidly-cued stimuli

Treisman and Gelade (1980) studied people's ability find the letter T in arrays of letters consisting of either Is and Ys, or Is and Zs. In the former condition, the T is distinctive from the other letters because it has a feature they do not---the horizontal line at the top, so people can simply search for this particular feature. By constrast, in the latter condition the T has no distinctive feature (i.e., both I and T have vertical lines, and both Z and T have horizontal lines), so people need to search for the combination/conjunction of the two distinctive features (i.e., the horizontal line together with a vertical line in the one letter). In their experiment, what effect did this difference between a feature search and a conjunction search have on the time taken to find the target T?

• People were roughly equally fast to find the T in both conditions when the array consisted of only a few letters; but as the number of letters in the array increased, they become much slower in the conjunction search than they were in the feature search condition.

• People were always slower to find the T in the conjunction search condition than in the feature search condition, and the difference between the conditions was consistent irrespective of how many letters the arrays contained.

• People took roughly the same amount of time to find the T, irrespective of whether they were performing feature search or conjunction search and irrespective of how many letters the arrays contained.

Why does an account of feature binding based on Hebbian learn require that features be bound into objects sequentially (i.e., one object at a time) for the process to be successfully completed?

• Because research using sequential-simultaneous paradigm shows that visual objects are encoded equally well when presented together in an array as when they are presented to one after the other

• Because if features from multiple objects are activated at once, they will be incorrectly bound together

• Because it takes a relatively long time to bind features into objects, so if extra features are activated unnecessarily the process is far less efficient