Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Comparative Approaches > How Vision Guides Action > Flashcards

How Vision Guides Action Flashcards

1
Q

ROBOTS VS AI?

A

ROBOTS:
- programmable; specific aim
- sensors & actuators to interact w/physical world
- autonomous/semi-autonomous
AI:
- algorithms that tackle range of issues ie. perception/navigation/logical reasoning
AI ROBOTS:
- ie. drone using autonomous navigation system to travel to goal location

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

BEETZ ET AL. (2011)

A
  • robotic roommates making pancakes
  • making sense of world (task/context) & making decisions (planning in robotics)
  • action sequence = interacting w/world (actuation & motion)
  • learning concepts from nature & applying them to AI systems & designs aka. bioinspired engineering
  • taking autonomous robot control from pick/place tasks to everyday object manipulation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

EWERT (1987)

A
  • neuroethology of releasing mechanisms: prey-catching in toads
  • simplest hypothesis = sensorimotor pathways for each action
  • each beh segment is mediated by separate releasing mechanism (RM)
  • motivation can modulate each RM
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

SIMMONS & YOUNG (1999)

A
  • toads respond to simple artificial stimuli
  • toad sits in glass vessel (prevents feedback from interacting w/stimulus); presented w/1/3 shapes (worm; antiworm; square)
  • prey model (P) = moved around it
  • response = number of turns to follow model in 1min
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

CAREW (2000)

A
  • invariant prey recognition by toad
  • configural stimulus properties elicit prey-catching response
  • stimulus variations don’t affect response
  • movement direction doesn’t affect response
  • velocity = crucial for worm stimuli BUT not for antiworm (dashed line)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

TOAD VISUAL PATHWAYS

A
  • typical vertebrate eye w/retina containing ganglion cells sensitive to edges; have centre-surround receptive fields
  • ganglion cells project retinotopically to optic tectum
  • optic tectum contains layers w/T5 cells which respond to moving stimuli
  • some ganglion cells project to thalamic-pretectal (TP) area
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

SUPERIOR COLLICULUS IN MAMMALIAN MIDBRAIN

A
  • homologous to optic tectum in vertebrate midbrain
  • tectal systems directs egocentric behavioural responses
  • neurons receive retinotopic visual input
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

MAIN SENSORI-MOTOR PROJECTIONS INVOLVED IN TOAD PREY-CATCHING BEH

A
  • sensory pathway = retina to thalamic-pretectal area (TP) & tectum (feature coding & identification of prey items)
  • motor pathway = tectum & TP to hindbrain (execution of beh response if prey = identified)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

SIMMONS & YOUNG (1999)

A
  • feature detectors in TP
  • TH3 cells in thalamic-pretectal area (TP) respond to moving visual stimuli
  • T5 cells in tectum belong to several cell classes; some respond to moving stimuli
  • T5(2) neurons exhibit stimulus invariant responses
  • T5(2) cells receive inhibitory input from T3 cells
  • experiments w/2 electrodes; 1 recording in T5(2); other stimulating TP neurons (T3 cells)
  • T3 cells not stimulated = T5(2) responded to stimulus; T3 cells excited = T5(2) cells became inactive
  • excitatory input = received by T3 cells from tectal cells; toads would either move towards/away stimulus
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

TOAD MODEL SUMMARY

A
  • T5(1) excited T5(2)
  • TH3 inhibits T5(2)
  • proof of causality = lesions in TP area change T5(2) activity & beh responses
  • control: worm > square > anti-worm
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

FINGERLING, EWERT, MENZEL & PFEIFFER (1993)

A
  • from toad -> robot; implementation of neurobiological principles of object discrimination in neural engineering
  • aka. pick-and-place robot w/toad’s prey-recognition system
  • alternative solutions can be found when considering how/why brains & nervous systems have evolved in dif ways in nature
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

BEH ORGANISATION: TRADITIONAL THEORIES

A
  • presume central controllers for beh organisation aka:
    1. NS structure (neuroreductionist view)
    2. internal representations (cognitivists view)
    3. contingencies presented by environment (behaviourists view)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

BEH ORGANISATION: GIBSON (1979)

A
  • critique & alternative conceptual approaches
  • locomotion & manipulation controlled not by brain but by info; control lies in animal environment system
  • rules governing beh aren’t like laws enforced by authority/decisions made by commander; beh = regular w/o being regulated
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

BEH ORGANISATION: WARREN (2006)

A
  • perception & action
  • dynamic systems coupled & independent
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

BEH ORGANISATION: LOBO ET AL. (2018)

A
  • ecological psychology
  • aka. perception & developmental psychology
  • “embodied, situated, non-representationalist approach to cognition”
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

PERCEPTION-ACTION CYCLE

A
  1. environment varies in predictable/unpredictable ways & at dif rates
  2. making sense of world w/sensory info/memory/uncertainties & deciding whether/which action to perform
  3. physiological/developmental/mental states vary/determine capacity to perform action
  4. world changes as consequence of action
17
Q

LAND ET AL. (1999)

A
  • eye fixations during tea-making
  • vision = active process; seeing/looking (gaze fixation)/sampling of info in time/space
  • primates/human eyes = highly mobile; eye movements can be slow (compensating shifts of gaze/foveal tracking) OR fast saccades (relocating gaze/fixating new target)
  • 1/3 fixations linked to subsequent actions (first fixations -> new objects)
  • 2/3 fixations after action (locating/directing/guiding/checking)
  • some fixations for locating weren’t followed by immediate actions; suggests some form of transaccadic memory exist as info isn’t lost when another saccade is made
18
Q

LAND & MCLEOD (2000)

A
  • from eye movements to actions aka. how batsmen hit the ball
  • ball leaves bowling machine at medium pace (60m/h)
  • gaze initially tracks ball; jumps to predicted location according to predicted ball trajectory (anticipatory saccade; good players duration only 130ms)
19
Q

LAND MF (2009)

A
  • events involved in an object related action as building block of task/goal directed action sequences
  • not only does visual system locate/recognise objects; ALSO continuously guides actions in order to produce adaptive beh responses
20
Q

VISUAL SYSTEM MODEL

A

SCHEMA SYSTEM
- select/supply next schema
GAZE SYSTEM
- find remembered object location & direct body/gaze
VISUAL SYSTEM
- 1. identify object; 2. monitor action
MOTOR SYSTEM
- 1. move trunk; 2. move hands to object; 3. perform action

Comparative Approaches (10 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions