Periodic Reporting for period 1 - MOPIGRASFIT (Motion Picture Grammatical Rating System for Infants and Toddlers)
Periodo di rendicontazione: 2015-07-01 al 2017-05-31
Existing motion picture rating systems in the EU only target viewers above age three although children become regular viewers at younger ages. Existing rating systems also classify motion pictures in terms of only their content although films differ substantially from conditions of natural perception and understanding films requires a film literacy. The overall objective of the proposed research was to assess systematically the role of: (a) natural perception, (b) cinematic techniques, (c) viewing experience and (d) cognitive development on the understanding of moving images. To this aim, three studies were conducted within the project. The first study was designed to see if babies can perceive the continuity of an action through the film cuts. It was already known that 12-month-old babies can predict other’s simple actions by performing goal-directed, anticipatory eye movements. However the studies testing this ability were using single run videos unlike the ones on TV. For this study, we recorded the actions from different camera angles and edited them together according to match-on-action editing technique, which is used by filmmakers to make edits unnoticeable. The results were surprising: 12 and 18-month-old babies were even better at action prediction when the actions were presented in multiple film shots. The second study was also designed on the basis of two concepts, one of which is from the developmental psychology (i.e. gaze following), and the other from the film studies (i.e. eye-line match). The eye-line match between two juxtaposed shots is based on the premise that an audience will want to see what the character on-screen is looking at. A film sequence with an eye-line match begins with a character looking at something off-screen, followed by a cut of another object or person. From a developmental perspective, it is well established that young infants turn in the direction that an adult has turned. The question was whether babies can still follow other’s gaze when the gazed object is presented in another film shot. The looking score (total of the correct looks, incorrect looks, and non-look) was significantly different between editing types (single vs multiple) and (babies vs adults). Babies at 12 and 18 month of age could successfully follow gaze in the unedited version but were less successful across edits. The third study was again designed on the basis of well-established concepts in developmental psychology (occlusion perception) and film studies (directional continuity). The studies on the developmental origins of occlusion perception show that by at least 6 months of age, infants are able to make anticipatory eye movements by fixating the leading edge of the occluder before an object reappears. It was not known however if babies can anticipate the right side of the screen when a horizontal movement is presented in two film shots. The results show that babies do not perceive the film cuts as occluders and can perceive the scene as a whole although it is presented from different camera angles because of the ongoing movement through the cut.
At the beginning of the project, the beneficiary took developmental psychology and advanced research method courses. Since she comes from film studies background, these courses helped her to broaden her horizons in general and design her studies in particular. She also attended eye tracking workshops and wide range of seminars on developmental psychology, observed infant studies going on at the Babylab . The beneficiary designed the studies and produced the stimuli. After the pilot studies, the stimuli were adjusted and testing was started. The beneficiary recruited the adult and infant participants. The initial findings of the experiments done with 12 month old babies were presented in 2017 at the The Society for Cognitive Studies of the Moving Image Conference, in Helsinki. The final results were presented in 2018 at Expanding the Field: Multi-Disciplinary Developmental Dynamics Conference in Norwich, UK and International Congress of Infant Studies in Philadelphia, USA. A book chapter on development of film literacy was also published within the project.
The findings of these studies have implications for our understanding of the cognitive neuro-developmental origins of film perception and an impact on the future of children’s film production. Together with the previous works done with babies the findings of this project show that babies as young as 6 month-age can perceive the similarities between a 2D image and the real 3D entities however before 18 month of age they struggle to understand that the 2D objects on TV are not real, for example they try to grasp the objects on the screen. Children younger than 18-month-of-age pay relatively high attention to salient formal features of TV rather than its content. By 12-month-of age they can follow an object moving across film shots horizontally when such shots are edited according to directional continuity rules. 12-month-olds can predict other’s simple actions when presented in edited sequences using match-on-action technique. Close-up shots help babies to focus on the relevant parts of the screen and this help them to understand the actions presented. However film-makers need to be cautious when editing the infant-directed videos and refrain unnecessary cuts. For example, using a close-up shot without an on-going action might cause a confusion for babies. It was observed that even pre-schoolers perceive the close-up shot of an unfamiliar film character as another and bigger character. By 12-month-of age babies use televised emotional reactions to guide their own behaviour by, for example, avoiding the toy the televised person didn't like. However they can't imitate one step actions when presented on TV before 15 months of age. Video deficit (better learning from in person than on video) peaks between 15 to 30 month-of age. Children between 15 to 36 month of age can only learn from TV only with repetition & parental support. By 18 month-of age attentional shift from salient to non-salient and content features occurs. They can differentiate standard coherent shot sequences from random shot sequences. Children at this age can imitate multi-steps simple actions presented on the TV. It is however not before 30 months of age that they can preserve multiple forms of input such as music, visual stimuli, and language when they are presented simultaneously. And it is not before 36 month of age that they can comprehend simple edited video stories as well as they comprehend unedited stories. Three-year-old children but not youngers can find a toy when they have watched it being hidden via screen. They comprehend time gaps in familiar actions and events when they are 4 year-old. Children younger than 4 year-old have also difficulties to understand that two actions presented in turn (cross-cutting) happen simultaneously. Four-year old children can not make sense of split screens. Only the ones with good working memory can make sense of horizontal camera movements(pans). Even 4 year-old children can't understand that cartoons are fictional. Fully developed ability to distinguish reliably between reality and fiction occurs later in childhood. Before they are 4 year-old, they also cannot understand the flashbacks, subjective camera angles.