The Visual Brain Part II

The Visual Brain Part II

1. CHALLENGES

   • COMMUNICATION
     • Issues in preverbal infants with verbal instructions, and equally, verbal response (Atkinson, 2002). Design should find novel ways of eliciting response from infant. This also ties in with consent.
   • COGNITIVE AND MOTOR ABILITIES
     • Limited set of behavioural responses due to stage of development; however, can use reaching & grasping as a response, as well as saccades & fixations (a relatively developed ability - Atkinson, 2002). It should be noted that for infants it could be hard to discern between cognitive deficits and visual problems.
   • ETHICAL CONSIDERATIONS
     • Cannot give verbal consent to partake in any experiment. Therefore, there should be no objective risk or discomfort, and guardians should be completely informed and reassured as to each stage. Moreover, motivational manipulations would be unethical.
   • MOOD AND STATE
     • Very changeable so design must take this into account- easy & varied. Prechtl (1978) devised a 5 point scale from deep sleep to crying- 3 ideal state for experimental testing.

2. TECHNIQUES

   • HABITUATION
     • Repeated presentation of a stimulus- attends to it less and less. A novel stimulus is then presented after a certain number of trials (point of habituation- 50% of initial looking time) = looking time increases, dishabituation. Simpler stimuli are habituated faster than complex. Indicates infant can see difference. Decrement in attention is due to evolution & learning (Atkinson, 2002). Bornstein et al. (1988) found similar results cross-culturally..
   • PREFERENTIAL LOOKING
     • Time spent/instances of fixations - looking at a particular stimulus. E.g. Fantz et al. (1962) stimuli with contours vs homogenous block. FPL popularised by Teller (1979) - limited looking options. + fixation point * simultaneous presentation Used in a lot of seminal work for visual acuity- e.g. fineness of gratings, will start to look more like a homogenous block- psychophysical staircase procedure. Recorded by observer or camera.
       • Fixations and saccades are not always fully developed- reports of hypotonic saccades (just fall short) under certain conditions (Aslin & Salapetek, 1975). Alternatively, 3D and 2D preferential reaching may be utilised (Atkinson & Braddick, 1998).
   • VISUAL-EVOKED POTENTIALS (VEPs)
     • Electrodes attached to scalp and neural activity recorded- non-invasive. Uses functional neuroanatomy to determine if the infant is processing the stimuli. VEP should be evoked at same frequency of visual stimulus.

3. Visual Development

4. STIMULI

   • DESIGNER STIMULI
     • Most research favours this approach as it deals with the target orthography instead of working retrospectively. These are stimuli that are designed to elicit a response from a particular region of the brain that is thought to be associated with a particular visual function. For example, gratings, orientation bars, rotations, luminance.
   • MARKER TASKS
     • Using adult neuropsychological populations and brain damage, as well as animal lesioning, performance and area functions in certain tasks- infer development by prescribing tasks to infants (Richards & Hunter, 2002).

5. ORIENTATION PERCEPTION: FUNCTIONAL ONSET

   • In their model, Hubel & Wiesel (1974) proposed the functional processing of orientation perception occurs in area V1. Braddick et al . (1986) - WHEN - VEPs. Infants presented with gratings- 24 images per second. Per frame, 1 x phase shift, every 3 frames, 1 x orientation shift 90 degrees. VEPs in 6-8 week olds were significant and sychronised with the orientation shifts & reversals. Phase shifts were used to make sure infants were not simply responding to the shift in luminance distribution brought about by an orientation reversal.
     • Later research used habituation techniques instead to rule out possibilities of VEP reaction to a property other than orientation (Atkinson et al., 1988; Slater et al., 1988). Habituate to static grating of either 45 or 135 degrees, then presented with same stimulus and flipped stimulus (novel)- more time spent looking at novel stimulus. However, this only demonstrated very coarse discrimination.
       • To address the question of sensitivity, Atkinson et al. (yr) used smaller orientations to determine the smallest discernible angle. Very fine (10) orientations were perceived poorly until around 12 weeks but 20 degrees was detected by around 45% of infants as young as 4 weeks. All orientation perception improved with age but 10 experienced the most drastic change.
       • Temporal frequency is also a function of orientation sensitivity. Atkinson et al. (1990) presented 3 reversals/sec and 8/ sec - significant VEP recordings for the lower temporal frequency- 3 weeks. Before 3 weeks, phase reversal responses were higher.

6. ILLUSORY CONTOUR PERCEPTION

   • Kanitza figures (Kanitza, 1979): 1) appear to be brighter than surrounds 2) sharp border perceived 3) appear to be closer than inducers.
     • WHERE: Von der Heydt, Peterhans & Baugartner (1984) performed single cell recordings for 'real luminance' defined versus illusory contours. Fixate at a point and stimulus was swept across visual field of neuron. Illuminated rectangle sweeps across visual field- increase of firing in V1 and V2. 2nd test- cut out in horizontal black bars at the top and bottom- perceive an illusory edge as they move together. V2 neurons fired but V1 neurons did not which suggests encoding doesn't occur until the V2.
       • To control for possible errors- e.g. response to another property - repeated but with cut-outs moving at different times- no unique V2 response, therefore not just responding to edges.
         • Experiment repeated again- when illusory contour was closed off with black edges, no neurons fired. Also presented at an angle to check if it was response to orientation.
           • Bertenthal, Campos & Haith (1980) used a habituation procedure instead. In first condition, habituated to Kanitza square and then shown 2 diff trials with different rotations of each part of Kanitza square. In second condition, habituated to non-Kanitza square, shown 2 trials again, one of which was a K square. If illusory square cannot be processed, will not become dishabituated between Non-K & K. (should be greater dishabituation). 7m better than 5m (still could at 5m).
             • *Criticism: illusory stimulus only one where all edges were facing inwards- could have influenced response- less likely but still possible.
   • Johnson & Aslin (1998) study of amodal completion - perceiving separate units, two rods not one (partly obscured). Ability to perceive illusory contours. Trial 1: occluded box and rod- no depth information. Trial 2: textures moving in different directions. In both, rods then move as two different ones- if they perceived trial 2 illusory contour, will dishabituate (4m olds). MOTION-DEFINED CONTOUR. Illusory contours just as useful as motion in providing depth ordering info.
     • Johnson & Mason (2002): can perceive shape using kinetic movements- when images were still on textured background- completely camoflauged.
   • Found that 2m olds preferred looking at side where illusory contour is placed (Rieth & Sirenteanu, 1993). However, infants could have just preferred the region with the most end-points. Therefore- Sirenteanu (1997) retested with 4-5m olds, can differentiate between vertical and oblique (diag) contours- not simply end points (however not same age group).
     • Curran et al. (1999) 8-22w olds. FPL- central fixation. Wave line ends form illusory contour, appears to move, versus real luminance-defined contour- more looking at ages 12-22. Due to line density? -> repeated with same line density in both conditions and sides- still preference. If done slowly enough even 2m olds could detect. Due to preference for alignment? Repeated with alignments and misaligned 'beads' on the lines. Showed preference for misaligned.
       • Otsuka & Yamaguchi (2003) moving vs static Kanitza square & non-K square. STATIC: 3-6m = no significant difference in looking time; 7-8m = more time at illusory figure. MOTION: all age groups spent more time looking at illusory figure. Motion plays a role.
         • Kavsek & Tonas (2006) - most studies confounded- preference could be due to perceived enhanced brightness. Therefore used inducers- opposite polarities (black and white). Rotations of inducers- preferred trial where square is induced.

Bilder und Dateien der Mindmap