Task 8 Flashcards
8.1: Cisek’s Affordance Competition view of brain functioning –> action selection
- natural environment presents animals with many opportunities and demands for action all the time e.g. food –> satiate hunger; predator–> caution/evasion
- cannot perform all behaviours simultaneously ==> action selection
8.1: Cisek’s Affordance Competition view of brain functioning –> action specification
animal must tailor the actions it performs to the environment in which it is situated e.g., grasping a fruit –> accurate guidance of the hand to the location of the fruit ==> action specification
8.1: Cisek’s Affordance Competition view of brain functioning –> how do action selection & specification interact?
processes occur continuously & simultaneously, in an ongoing competition for the currently most relevant action
8.1: Cisek’s Affordance Competition view of brain functioning –> What are ‘affordances’?
constant competition between internal representations of the potential actions
8.1: Why does Cisek think that the parietal cortex plays an essential role in this ongoing competition?
parietal cortical areas are strongly & reciprocally interconnected with frontal regions involved in movement control –> LIP is interconnected with FEF (–> saccades), MIP with PMd & primary motor cortex (M1) (–> limbs), AIP with ventral premotor cortex (PMv)
8.1: Where does Cisek think the competition between potential actions play out?
mostly within reciprocally interconnected fronto-parietal system
8.1: How does the PFC via attentional modulation bias the ongoing competition?
-along the dorsal stream –> increasing influence of attentional modulation, with information from particular regions of interest enhanced while information from other regions is suppressed
—> parietal representation of external space becomes increasingly sparse as one moves away from striate cortex
–> only the most promising targets for movements make it so far to be represented in the parietal cortex
==> competition is biased by input from BG & PFC regions which collect information for action selection
8.1: What does Cisek mean by saying “Selective attention is seen as an early mechanism for action selection”?
- Because only the most promising targets for movements make it so far to be represented in the parietal cortex
- reduces the volume of information that is transformed into action-related representations
8.2: What is Cisek’s (2007) view of the functional organization of the parietal “affordance” path towards action (i.e., from the parietal cortex to the frontal cortex)?
- begins in visual cortex & proceed rightward across the parietal lobe, transforming visual information into representations of potential actions
- three neural populations along thisroute:
- -> leftmost: encoding of potential visual targets, modulated by attentional selection
- middle: potential actions encoded in parietal cortex
- rightmost: activity in premotor regions
- each population is depicted as a map of neural activity, with activity peaks corresponding to the lightest regions
- as action specification occurs across the fronto-parietal cortex, distinct potential actions compete for further processing
- competition is biased by input from BG & PFC regions which collect information for action selection
- PFC: sends votes
- BG: value-based DM
8.2: Compare Cisek’s view with Binkofsky & Bixbaum from task 1- substreams
- B&B: only 2 substreams
- Cisek: more than 2 substreams
8.2: Compare Cisek’s view with Binkofsky &Bixbaum from task 1- attention
- B&B: not much influence of early attention
- Cisek: attention influences which actions are represented –> early attention
8.2: Compare Cisek’s view with Binkofsky & Bixbaum from task 1- dependency on movement/effector?
- B&B: substreams involve different brain regions but these are NOT dependent on effector
- Cisek: different substreams comprise different brain regions depending on movement/effector
8.2: Compare Cisek’s view with Binkofsky & Bixbaum from task 1- competition & bias signals
- B&B: competition between streams is possible but this is NOT influenced by bias signals
- Cisek: competition between multiple potential actions that is resolved by bias signals (from BG & PFC)
8.2: Compare Cisek’s view with Binkofsky & Bixbaum from task 1- feedback
- B&B: no role of feedback
- Cisek: overt feedback (from environment) & internal feedback (from cerebellum) will influence future potential action decisions
8.2: Compare Cisek’s view with Binkofsky & Bixbaum from task 1- serial or parallel?
- B&B: relatively serial view of processing
- Cisek: more parallel processing
8.3: Where in the brain do decisions take place according to Cisek’s view?
- made within the very same neural circuits that control the execution of those actions
- Fronto-parietal circuit (PMd & MIP) –> DM based on visually guided reaching
- -> Uses info from PFC
- if a decision begins to emerge in one region, then it will propagate outward to other regions
- DM through distributed consensus
- -> accumulation of evidence in favour of given choice
8.3: Where in the brain do decisions take place according to Cisek’s view? –> example (ideomotor vs ideational apraxia)
- Ideomotor apraxia: impairments of the implementation of these concepts
- Ideational apraxia: impairments of the conceptual representation of actions
- Ideomotor apraxia –> dorsal networks for action specification
- Ideational apraxia –> ventral networks for action selection
8.3: Where in the brain do decisions take place according to Cisek’s view? –> example with ideational apraxia
- predict deficits in object use due to problems in ‘‘affordances’’ triggering appropriate actions
- may thus arise due to wrong actions being generated according to errors in affordances
- E.g., object may be recognized for its use, yet present the actor with graspable features (affordances) that may be similar to other objects (e.g., grasping a toothbrush may be similar to grasping a knife) and lead to activation of subsequent action representations that are inappropriate for the object at hand
- unable to select from these competing actions (or inhibit them)
8.3: Where in the brain do decisions take place according to Cisek’s view? –> relate to flow chart (task 0)
- in all loops/ pathways
- -> competition, but in end all need to be in consensus in order to make decision (threshold idea) (finding a compromise) –> in the end one action is decided for
- depends on context; bias signals from BG& PFC
- depends on motivation, cognition, motor…
- flow chart: all steps in the middle
8.4: Describe the filter-view on the dorsal attention system & their attention mechanisms
- traditional filter view (dorsal attention system): attention as regulator of sensory representations –> limited subset of sensory signals reach processing stages
-attention mechanisms lead to a selection of only a small number of possible objects being represented
neutrally
8.4: Introduce the notion of priority maps (see the flow chart in task 1) (Krauzlis et al., 2014).
-KRAUZLIS: dorsal attention system sets priorities for processing particular stimuli over others –> these priorities stem from the current task that a person is engaged in (i.e. the dominant state)
8.4: Explain how this view is only half of the story; that taking a wider perspective leads to the insight that these priority maps serve the larger good of goal-directed behavior –> simple example (for instance the grocery problem in task 0)
- change due to external circumstances
- priority maps serve the larger good of goal-directed behaviour
- e.g. you usually prefer bananas BUT if bananas are green or if you’ve eaten them a week in a row –> you’d decide for pineapple
- -> SO priority maps change due to external circumstances (motivation, cognition, etc)
8.4: Explain the paper title “Attention as an effect not a cause” (Krauzlis et al., 2014).
- Attention: effect of interpreting sensory data
-consequence of circuits centered on BG involved in value-based motor & non-motor DM
-competition between possible interpretations of current state by BG rather than competition to determine how sensory data is represented in neocortex
-states differ in weights they assign to various inputs
-filter like properties associated with attention result from particular weights applied to inputs that define current state
==> FILTER LEADS TO ATTENTION! (Cisek& Krauzlis)
–> Cisek: fronto-parietal pathway/network filtering leads to attention (more abstract form of DM)
-BG also plays a role; integrates both movement parameters & decision variables such as reward
-PFC: decision making (advanced action selection)
–> Krauzlis: basal ganglia state estimation leads to attention (different states; if attention shifts, state shifts) –> Krauzlis: competition in BG
==> ATTENTION IS NOT A CAUSAL CAUSE
8.5: Explain the concept of a state (Krauzlis et al., 2014).
=current condition of a subject in its environment
8.5: How is the concept of a state used in the paper to give attention to its correct position in the larger process of goal-directed decision making? (Krauzlis et al., 2014)
- each state = template for particular situation & dedicates what the relevant stimuli, actions & rules are
- specific stimuli attended = specific state
- involves interpreting different info (e.g features of external world, internal status of subject)
8.5: Why do the authors think that the BG are essential in the competition between states? (Krauzlis et al., 2014)
- competition between possible interpretations of current states
- learn associations between states & set of actions with the highest expected reward (reward –> dopamine)
- receive info from many different areas (input) & weights inputs according to the current state
- if state changes, attention shifts!!!
8.6: Similarities between the views of Cisek & of Krauzlis et al. –> Competition
- -> Cisek: internal representations of potential actions are in constant competition
- competition in frontal-parietal loop
- competition is resolved by PFC& BG –> underestimates importance of BG
- -> Krauzlis: competition between possible interpretations of current state within BG –> stresses importance of BG
8.6: Similarities between the views of Cisek & of Krauzlis et al. –> Subordinate role of attention
Cisek:
- many possible actions are eliminated from processing by selective attention mechanism which limit sensory info that can be transformed into action
- attention results from a filter, letting through less and less info as info processing progresses from lower-levels to the higher-levels
- > attention results from filter/biasing input to change representations of sensory info (“votes”)
Krauzlis:
- attention arises as a result of competition of possible states in BG
- different sensory info & knowledge is unequally important between the different competing states -> their impact on perception & action is limited by which of those states is currently dominating
- -> attention not necessary to change representation of sensory info (representation of sensory info/state is relevant for chaning attention)
8.6: Similarities between the views of Cisek and of Krauzlis et al. –> Primacy of goals
Cisek:
- PFC & BG function as a gate for action execution
- key to this gate are certain decision variables - which decide which action is in line with goals
Krauzlis:
-BG estimates the most appropriate state for the current situation to decide which action is in line with current goals
Cisek –> subpathways of dorsal pathway (LIP & MIP): responsible for movements of what?
LIP –> saccades
MIP –> limbs
Krauzlis BG role of attention
- attention arises as a functional consequence of circuits in BG
- -> sensory input
- -> internal state
- -> state estimation
- -> decision policy
8.6: Similarities between the views of Cisek & of Krauzlis et al. –> Priority maps
- Cisek: under the influence of attentional modulations & biases, only the most promising targets for movements make it so far to be represented in the parietal cortex
- Krauzlis: dorsal attention system sets priorities for processing particular stimuli over others ->these priorities stem from the current task that a person is engaged in (i.e., the dominant state)
