Here we present a protocol for familiarization-test paradigms which provide a direct test of infant categorization and help to define the role of language in early category learning.
This familiarization test method reveals how infants form visual object categories and what kinds of conditions most powerfully support their learning. Using only infants eye movements we can measure how successfully they learn a category. Then we can directly compare that learning across a huge range of different learning conditions.
Begin by creating the visual stimuli for the experiment. To create a new continuous category, first design a pair of novel digital images. Then, morph the pair of images together to form a continuum of exemplars between the two original images.
Create at least two categories in this way, so that one can serve as the category to be learned, while the other provides the novel category exemplar for the test trial. Next, select the familiarization exemplars at evenly spaced intervals from across each learned categories continuum. Select an appropriate number of exemplars and commensurate with the difficulty of the category and age of the participants.
To create the exemplars for the test phase, select the midpoints of the familiar categories continuum and the novel categories continuum. Then match the color of the novel exemplar to that of the familiar exemplar using an image manipulation program. For both the labeling and non-labeling phrases use previously recorded auditory stimuli produced by a female native English speaker in a sound proof booth.
To collect adequate data for a familiarization test measure, most widely available eye trackers will suffice. The objects occupy large portions of the screen and the data analysis investigates performance over a long window. Using the eye tracker software, create four different conditions, fully supervised, unsupervised, semi-supervised and reversed semi-supervised.
Ensure these conditions are separate, so that each infant will see only one condition. Next, generate at least two pseudorandom orders of the learning exemplars with the constraint that no more than two exemplars from the same side of the continuum can be shown consecutively. Now use video editing software to create familiarization videos that pair the previously created auditory stimuli with the visual stimuli as appropriate for each condition.
For instance, in the fully supervised condition pair each familiarization exemplar with a labeling phrase. In the unsupervised condition, pair each familiarization exemplar with a non-labeling phrase. In the semi-supervised condition, pair only the first two exemplars in each order with labeling phrases, but the rest with non-labeling phrases.
For the reversed semi-supervised condition, pair the final two exemplars with labeling phrases but the first four with non-labeling phrases. Upload these videos into the eye tracker software being sure to order the familiarization videos as determined by the pseudorandomized order. Also upload a short attention grabbing animation displayed in the center of the screen after familiarization, to ensure that most infants look at the center of the screen when the test phase begins.
Finally, for each learning category, design two test trials each featuring two exemplars displayed side by side. For both, ensure that one exemplar represents the midpoint of the now familiar category, while the other represents the midpoint of the novel category. Ensure that test trials last 5 to 20 seconds to accumulate sufficient looking, and counterbalance the trials so that the left right positioning of the novel exemplar in the test trial is reversed across the videos.
Upload the task trials to the eye tracker software, positioning them after the post familiarization attention getter. Counterbalance these trials presentation, so each infant has an equal chance of seeing a left novel or right novel test trial. Before the infant arrives, setup the eye tracker.
Randomly assign the infant to a condition and an order, then open the eye tracker software, and select the assigned condition order pair, and enter the participant number for the recording. After obtaining consent for the study, bring the infant and the caregiver to the eye tracking room. Ensure this room is moderately lit without any distracting decorations on the walls.
Seat the caregiver at an appropriate distance for the model of eye tracker being used. If the infant does not wish to sit in the caregivers lap, they may sit on their own or they may sit in a car seat. If the infant is sitting on the caregivers lap, instruct the caregiver not to bias infants behavior in any way, but to try to keep the infant centered on the caregivers lap.
Then, provide the caregiver with a pair of blacked out sunglasses to wear, so they cannot see the stimuli. Now ask the infant to look at the eye tracker screen. Consider displaying an engaging image or video to attract their attention.
The screen should be positioned, so that the infant eyes are within the calibration window. Next, perform the eye trackers calibration procedure using a five point calibration, if possible. Infants often respond better when the calibration image is animation with auditory accompaniment.
If the infant passes calibration, then begin the experiment if not, recalibrate until they are successful. Any infants who cannot be calibrated should be excluded from the study. This figure shows mean preference scores across conditions.
Infants in the fully supervised and semi-supervised conditions, displayed novelty preferences significantly above chance. While infants in the unsupervised and reversed semi-supervised conditions, performed at chance levels. Here we see the infants looking patterns during testing.
In the fully supervised and unsupervised conditions and in the semi-supervised and reversed semi-supervised conditions, infants pattern of looking to the exemplars diverged between 3450 and 3850 milliseconds. The colored shaded regions indicates standard error of the mean. This procedure can be used with infants of various ages and materials that vary in complexity.
As a result, each new design must reflect infants learning capacity at that intended age. We can go on to ask, whether and how infants capacity as measured here, correlates with or even predicts infants acquisition of later language milestones. Such as vocabulary size or word order.
This technique has allowed developmental scientist to study cognition and perception, in infants as young as two months. It continues to illuminate the ways that language facilitates learning throughout infancy.
