Limits...
Visual motherese? Signal-to-noise ratios in toddler-directed television.

Wass SV, Smith TJ - Dev Sci (2014)

Bottom Line: We compared toddler-directed and adult-directed TV programmes (TotTV/ATV).Our results suggest that TotTV designers have intuited techniques for controlling toddler attention using low-level visual cues.The implications of these findings for structuring childhood learning experiences away from a screen are discussed.

View Article: PubMed Central - PubMed

Affiliation: Medical Research Council Cognition and Brain Sciences Unit, Cambridge, UK.

Show MeSH

Related in: MedlinePlus

Comparing feature congestion between our TotTV and ATV-Live samples. Almost all of the faces in our TotTV programmes were either CGA, Cel or live action puppets, whereas all the faces in the most popular ATV excerpts were of human actors. For each example, the image on the left shows the input image, identified via hand-coding; the image on the right shows the feature congestion map; light-coloured areas are more congested. The number to the right shows the average feature congestion over the whole face.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4309493&req=5

fig07: Comparing feature congestion between our TotTV and ATV-Live samples. Almost all of the faces in our TotTV programmes were either CGA, Cel or live action puppets, whereas all the faces in the most popular ATV excerpts were of human actors. For each example, the image on the left shows the input image, identified via hand-coding; the image on the right shows the feature congestion map; light-coloured areas are more congested. The number to the right shows the average feature congestion over the whole face.

Mentions: Figure 7 shows a sample of character faces analysed for feature congestion. Total feature congestion across the whole face is higher in the TotTV than the ATV samples; also interesting, however, is that in the TotTV samples the areas of highest feature congestion are located around the eyes, which are semantically the most informative areas within the face (Grossman & Johnson, 2007). This was true both for all animated characters (including ATV-Anim as well as TotTV) and for costumed live action TotTV characters. In the ATV-Live samples, in contrast, feature congestion around the eyes does not appear to be higher than it is in the rest of the face (Figure 7) and feature congestion in the face is not higher than in the rest of the frame (Figure 5f). Our findings suggest that feature congestion, which is a low-level exogenous influence on gaze location (Rosenholtz et al., 2007), may be used in TotTV to guide attention both to the semantically most important part of the frame (the face) and to the semantically most important part of the face (the eyes). One irony is that their simplified design and limited range of movement means that the eyes of cartoon/puppet characters may not have the full range of nuance and expression that real eyes do. However, for younger viewers this reduced complexity may accentuate critical expressive features and increase comprehensibility, which in turn leads to increased attention (Pempek, Kirkorian, Richards, Anderson, Lund & Stevens, 2010, Kidd, Piantadosi & Aslin, 2012).


Visual motherese? Signal-to-noise ratios in toddler-directed television.

Wass SV, Smith TJ - Dev Sci (2014)

Comparing feature congestion between our TotTV and ATV-Live samples. Almost all of the faces in our TotTV programmes were either CGA, Cel or live action puppets, whereas all the faces in the most popular ATV excerpts were of human actors. For each example, the image on the left shows the input image, identified via hand-coding; the image on the right shows the feature congestion map; light-coloured areas are more congested. The number to the right shows the average feature congestion over the whole face.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4309493&req=5

fig07: Comparing feature congestion between our TotTV and ATV-Live samples. Almost all of the faces in our TotTV programmes were either CGA, Cel or live action puppets, whereas all the faces in the most popular ATV excerpts were of human actors. For each example, the image on the left shows the input image, identified via hand-coding; the image on the right shows the feature congestion map; light-coloured areas are more congested. The number to the right shows the average feature congestion over the whole face.
Mentions: Figure 7 shows a sample of character faces analysed for feature congestion. Total feature congestion across the whole face is higher in the TotTV than the ATV samples; also interesting, however, is that in the TotTV samples the areas of highest feature congestion are located around the eyes, which are semantically the most informative areas within the face (Grossman & Johnson, 2007). This was true both for all animated characters (including ATV-Anim as well as TotTV) and for costumed live action TotTV characters. In the ATV-Live samples, in contrast, feature congestion around the eyes does not appear to be higher than it is in the rest of the face (Figure 7) and feature congestion in the face is not higher than in the rest of the frame (Figure 5f). Our findings suggest that feature congestion, which is a low-level exogenous influence on gaze location (Rosenholtz et al., 2007), may be used in TotTV to guide attention both to the semantically most important part of the frame (the face) and to the semantically most important part of the face (the eyes). One irony is that their simplified design and limited range of movement means that the eyes of cartoon/puppet characters may not have the full range of nuance and expression that real eyes do. However, for younger viewers this reduced complexity may accentuate critical expressive features and increase comprehensibility, which in turn leads to increased attention (Pempek, Kirkorian, Richards, Anderson, Lund & Stevens, 2010, Kidd, Piantadosi & Aslin, 2012).

Bottom Line: We compared toddler-directed and adult-directed TV programmes (TotTV/ATV).Our results suggest that TotTV designers have intuited techniques for controlling toddler attention using low-level visual cues.The implications of these findings for structuring childhood learning experiences away from a screen are discussed.

View Article: PubMed Central - PubMed

Affiliation: Medical Research Council Cognition and Brain Sciences Unit, Cambridge, UK.

Show MeSH
Related in: MedlinePlus