Visual Memory & Attention PDF

PY40008 & PY50408 INFANT DEVELOPMENT Visual Memory & Attention “It is no exaggeration to say that without looking time measures, we would know very little about nearly any aspect of infant development” Aslin, 2007, Dev Sci, 10, 648-53  Looking Time...

PY40008 & PY50408 INFANT DEVELOPMENT Visual Memory & Attention “It is no exaggeration to say that without looking time measures, we would know very little about nearly any aspect of infant development” Aslin, 2007, Dev Sci, 10, 648-53  Looking Time Measures Provide Information About  Visual memory  Attention  Individual differences  Prediction of IQ  Cognition (e.g. reasoning)  Knowledge (object properties, events) Aslin, 2007, Dev Sci, 10, 648-53  Main Topics Today  Habituation & habituation methods  Evidence for recognition memory  The comparator model  Speed of processing & disengagement hypotheses  Heart rate measures to identify phases of visual attention  Relation between looking time and childhood IQ  Infant visual preferences and stimulus complexity  Visual Habituation Fixation Time in Seconds 10  Stimulus presented Novelty for a series of trials 8  Look duration 6 decreases  Stimulus is eventually 4 recognised 2 Recognition  Habituation involves learning & memory 0 1 2 3 4 5 6 7 8 9 10 Trials Colombo, 1993, Infant cognition: Predicting later intellectual functioning, Chapter 2 Testing For Genuine Habituation  Reduced looking may occur for several reasons  Information processed and stored in memory  Change of state (sleepy)  General loss of interest  Test to see if looking time increases again for a novel stimulus (recovery)  Recovery or novelty preference indicates memory for the habituated stimulus  Testing For Genuine Habituation Recovery from habituation Total look duration 12 10 8 Novel 6 4 Significantly longer looking at the novel 2 Familiar stimulus indicates recovery from 0 habituation 1 2 3 4 5 Trials Visual Paired Comparison (VPC) Familiarisation Single or Paired Visual Paired Comparison (VPC) Novelty Preference Test Significantly longer looking at novel (>55%) indicates recognition Recording Infant Habituation Observer Picture on screen Recording Infant Habituation  Infant Control Procedure  Duration of exposure to the stimulus is controlled by the infant  Trial starts when infant looks at the stimulus  Trial ends when infant looks away for 2 sec (some researchers use 1.5 or 1 sec)  Trials continue until habituation criterion is reached  Baseline – average of first two or three trials  Criterion – average of two or three consecutive trials is 50% or less than baseline Horovitz et al, 1972, Dev Psychol, 7, 90  ICP - Rapid Habituation Total look duration 28 sec 12 Baseline: average of first 2 10 trials = 9.4 sec 8 6 Criterion in 4 trials 4 Average = 4.4 sec 46% of baseline 2 0 1 2 3 4 5 6 7 8 Trials  ICP - Slow Habituation Total look duration 60 sec 12 Baseline = 10.2 sec 10 79% 8 91% 72% 6 51% 4 34% 2 Criterion in 8 trials 0 1 2 3 4 5 6 7 8 Trials What Happens During Habituation?  Sokolov’s Comparator Model Habituation Response Cortical model Match Match Input Orienting Mismatch Response Autonomic Response Arousal system (Heart Rate) Sokolov, 1963, Perception And The Conditioned Reflex  Habituation Involves Both Learning & Memory Fixation Time in Seconds 10 Novelty  Stimulus presented 8 for a series of trials 6 Encoding &  Look duration Comparison decreases 4  Two main processes 2 Recognition  Encoding  Comparison 0 1 2 3 4 5 6 7 8 9 10 Trials Colombo, 1993, Infant cognition: Predicting later intellectual functioning, Chapter 2  The Comparator Model – An Information Processing Account  A mental representation is gradually constructed during familiarisation  Mismatch - arousal system activated, attention remains engaged, infant keeps looking  Match – arousal system and attention are inhibited, infant stops looking  Looking time gets progressively shorter as more information is stored and there is less to process  Looking time equates to processing time  Novelty preference indicates memory and recognition Rose et al, 2004, Dev Review, 24, 74-100 Individual Differences – Short & Long Lookers  Individual Differences In Look Duration Are Stable  Look duration is a relatively stable characteristic in infants  Short lookers remain short lookers; long lookers remain long lookers Colombo, 1993, Infant cognition: Predicting later intellectual functioning, Chapter 2, p.52 Explanations For Individual Differences In Looking Time  The Comparator Model supports the speed of processing hypothesis  Short lookers are faster processors  But there is an alternative  The disengagement hypothesis  Short lookers are faster at disengaging attention and looking away  Testing the Speed Hypothesis  Vary the duration of familiarisation in the VPC paradigm (short, medium, long)  Constant duration for the novelty-preference test  Faster processors will show a novelty preference after shorter familiarisation  Slower processors will show a novelty preference after longer familiarisation  A Quick Way To Identify Short & Long Lookers  View picture of a face for 20 sec  Record duration of each look  Identify the longest look  Split group at the median  Short lookers below median  Long lookers above median Colombo et al, 1995, Cog Dev, 10, 271-285 Speed Differences at 4 Months Novelty Preference Scores (Preference test = 10 sec) Familiarisation 10 sec 20-30 sec 40 sec Short lookers 68%* 64%* - Long lookers - 47% 67%* Short lookers are faster processors because they show a significant novelty preference after a shorter familiarisation Freeseman et al, 1993, Child Dev, 64, 1191-1203 Short Lookers Disengage Faster Disengagement Procedures at 3 and 4 Months Competition Non-competition condition condition Frick et al, 1999, Child Dev, 70, 537-548 Short Lookers Disengage Faster Disengagement Procedures at 3 and 4 Months Competition Non-competition condition condition Look duration & latency Look duration & latency r =.41, p <.01 r =.07, ns Frick et al, 1999, Child Dev, 70, 537-548  Looking Time Is Not The Same As Processing Time  Looking time includes  Time spent orienting and engaging attention  Time spent processing information  “Blank” looking while infant tries to disengage from the stimulus  You can’t tell what is happening simply by watching the infant  But there is another method … What Happens To Heart Rate When The Infant Looks At A Novel Stimulus? The comparator model also involves control of autonomic responses  Heart Rate And Phases Of Attention Look Infant Looking away 2 1 2 3 4 5 1. Baseline 2. Orienting (OR) Heart rate change (bpm) 0 3. Sustained attention (SA) -2 4. Attention termination (AT) -4 (Disengagement) 5. Return to baseline -6 Lower heart rate during -8 sustained attention -10 Heart rate recovers at end -5 0 5 10 15 20 25 of sustained attention Seconds following stimulus onset Richards & Casey, 1991, Psychophysiology, 28, 43-53  Heart Rate, Processing & Distractibility Look Infant Looking away 2 1 2 3 4 5 0 Heart rate change (bpm) -2 -4 -6 -8 -10 - 0 5 1 1 2 2 5 0 5 0 5 Seconds following stimulus onset Easy to distract Hard to distract Richards & Casey, 1991, Psychophysiology, 28, 43-53  Heart Rate And Processing  Three-month-olds were more easily distracted when heart rate accelerated, than when it decelerated (Richards, 1989)  Looking to a distractor occurred more at the start of a fixation (OR) and less during heart rate deceleration (Richards & Turner, 2001)  Heart rate deceleration indicates sustained attention and information processing Richards, 1989, Inf Beh Dev, 12, 425-436 Richards & Turner, 2001, Child Dev, 72, 963-972  Heart Rate And Information Processing  Stimuli were presented briefly (5 or 6 sec) during different phases of attention  Presentation during sustained attention produced a significant novelty preference  No evidence of recognition memory when stimuli were presented during attention termination Frick & Richards, 2001, Infancy, 2, 331-352 Richards, 1997, Dev Psychol, 33, 22-31  Short Lookers Are Faster At Both Processing Information And Disengaging Pretest – 20 sec Heart rate measures Familiarisation – 20 sec Heart rate measures  Each infant’s longest look (peak look) was identified during both pretest and familiarisation  Duration of Orienting, Sustained Attention and Attention Termination phases obtained from heart-rate recordings Colombo et al, 2001, Child Dev, 72, 1605-1616  Peak Look Duration is Related to Both Sustained Attention and Attention Termination  Peak look duration was unrelated to orientation duration (OR)  Infants with longer peak looks spent more time in both sustained attention (SA) and attention termination (AT) Average Peak Attentional phase Look Duration OR.11 SA.45*** AT.58*** ***p