loading

Logout succeed

Logout succeed. See you again!

ebook img

Category-specific attention for animals reflects ancestral priorities, not expertise PDF

pages21 Pages
release year2007
file size1.97 MB
languageEnglish

Preview Category-specific attention for animals reflects ancestral priorities, not expertise

Category‐specific attention for  animals reflects ancestral  priorities, not expertise Joshua New, Leda Cosmides, and  John Tooby Hao Ye VJC 10/18/2007 Visual Attention • some areas of the visual field are  selected for additional processing • selection based on: – current goals – personal relevance – low‐level features Origins • goal‐derived – aids in voluntary visual search – additional attention to task‐relevant objects • ancestrally derived – evolutionary benefit for involuntary additional  processing (environmental dangers) • expertise‐derived Animate Monitoring Hypothesis • ancestrally‐derived • automatic allocation of additional attention – humans – non‐human animals • important for: – food – danger (predators / human adversaries) – social opportunities – cues for other animals, plants, or humans,  environment Paradigm • change‐blindness task – subjects presented with color images of complex  scenes – each scene has two versions with a change • presented for 250ms • separated by 250ms blank mask – repeats until subject responds yes or no • changes occur in multiple semantic categories Timing Predictions • changes to animals (incl. humans) will be  detected more quickly • changes to animals (incl. humans) will be  detected more frequently (higher accuracy) • detection advantage is not due to interest  level of target • detection advantage is not due to low‐level  stimulus properties of target • detection advantage not a result of experience Experiments • experiment 1: – 70 scenes in 5 semantic categories {people,  animals, plants, moveable artifacts, fixed artifacts} • experiment 2 – duplicate of experiment 1 • experiment 3 – inverted stimuli – preserves low‐level properties – disrupts recognition Experiments • experiment 4 – Gaussian blur – preserves most low‐level properties – further disrupts recognition • experiment 5 – 96 scenes in 4 semantic categories {vehicles, fixed  artifacts, animals, people} Results • experiments 1 & 2 – animate (people and animals) targets were  detected faster than inanimate (plants, moveable  and fixed artifacts) targets [p = 10‐10, 10‐15] – animate targets were detected more often (higher  hit rates) than inanimate targets [p = 10‐8, 10‐10] • interest ratings correlate with RT, but not after  controlling for animacy

See more

The list of books you might like