Indices into IPS text and video materials.
LC3cV01 Video Control and Index

Video Corpus Structure and Detail

The video corpus was created between 1978 and 1984, initially as a collection of 30 minute reel-to-reel videotapes made with a Sony Port-a-Pac (owned by the MIT Artificial Intelligence Laboratory). The research objective was to explore the extent to which, in observing the everyday life of one child, one could trace in detail the interactions of everyday activities and the child’s developing understanding of her physical and social world. The corpus was constructed as a fusion of three epistemological perspectives, the structural focus of Piagetian development, the functional commitments of Artificial Intelligence, and the context appreciation of Ecological Psychology, brought together through commitment to Case Study within a family setting.

Over the intervening years. the videos have undergone format conversion: first from reel-to-reel to VHS tape, next to digitized DVD, then to digital mass storage. Through the past decade and more, the original videos, typically in 3 or 4 sessions of different foci, have been rendered as “panels” (each containing the clips of one videotape), segmented into theme focused “Panes”, uploaded here for online access. The grounds for segmenting the video tapes into themes and clips are revealed in the name codes of each clip. The number following the initial “P” is Peggy’s life-week. The letters represent divisions based on “Themes,” or a significant new initiative in the action. Segments are further subdivided (into sequentially numbered clips) to keep the digital file size small enough for easy network transfer.

The titles of clips shown in the individual panes and the text below them suggest what the activity is relevant to. Many specify the objects in the immediate context. The labels are not claims. For example, in P167F, where the title is “Juggling,” there is no intent to suggest competence or instruction. The intent is to suggest the activities may involve goals, pre-cursors or components of mature, later developing activities. Similarly, in the clips labeled “Discrete Substance,” one is not “teaching to” Piaget’s experimental task as any sort of test, but rather using his experimental situation for activities in which one might see and capture a path of incremental development for this child.

The current initiative is to unpack the video detail with two other representations of the content: transcribed text and video subtitles. The clips’ detail level can be reached through their contemporaneous Panels. With a thematic focus, clips can be directly accessed through Index located links. As the unpacking to transcriptions and analyses proceeds, more clip links will fill out the cells of the Index table.

The rows of the Index table are in sequence by serial day date since Peggy’s birth. The dates of the video sessions are less precise after Peggy’s fourth year begins. Our “experiments” typically were on Sundays, when we were all at home and had fewer outside pressures. If the actual day is known to be some other specific date, that is specified. Otherwise the dates mean “this date is the Sunday of Peggy’s Nth week. It would be unusual, perhaps even a surprise if the actual session date were different by more than one day or two.

The video distribution goes across the six years of the study but is concentrated in Peggy’s first three and a half years. Uploading further Panels to “fill in the gaps” will continue in conjunction with the unpacking initiative.

Names, IDs and representation details used in the transcription dialogues:

IDs appear as names in the Lawler text transcripts but not in the videos. Where actions are manifest in the video clips, though not marked by sounds, they are noted thus [descriptive text]. The action notes are commonly interspersed in the text of a dialogue with a different agent ID. To avoid confusion in such cases, a single letter identifying the agent maybe included. Similarly, when speech of vocal expressions of a second party are interspersed in a block of text with a different agent’s ID, that vocal expression is cited thus {P: “That” (dhat, or /∂aet/)}. I intend to represent vocalizations as potential words, eg. “That”, and as sounds according to the Wikipedia pronunciation guide (dhat or /∂æt/); ; limitation: when my keyboard does not permit accurate transcription of IPA symbols, their omission will be shown by /xxx/ or by a common name,eg, “short e”.
The episodes into which the transcripts are divided are made in the best judgment of ‘The Analyst,’ me.

The automatic transcriptions, provided by “Happy Scribe.com” are expected to be 85% correct as to words recognized. Their text and video editor permits word corrections, speaker identity corrections, insertion of action notes, and restructuring of speech and activity descriptions. The editorial extensions are of critical importance, especially as the primary subject of the video data is met as a pre-verbal infant. Once edited, I refer to these texts as “Lawler Text Transcriptions” and install them as the episodic “Clip notes” attached to the subtitled videos in the detailed level of transcription. Where alternate interpretations of behavior are desired or needed, the video may be directly consulted ad lib, either with subtitles at the detail level or without them, at the Panel level.

LC3cV01 Video Control and Index