Limits...
Competing visual flicker reveals attention-like rivalry in the fly brain.

van Swinderen B - Front Integr Neurosci (2012)

Bottom Line: There is increasing evidence that invertebrates such as flies display selective attention (van Swinderen, 2011a), although parallel processing of simultaneous cues remains difficult to demonstrate in such tiny brains.Local field potential (LFP) activity in the fly brain is associated with stimulus selection and suppression (van Swinderen and Greenspan, 2003; Tang and Juusola, 2010), like in other animals such as monkeys (Fries et al., 2001), suggesting that similar processes may be working to control attention in vastly different brains.Visual competition dynamics in the fly brain were dependent on the rate of pattern presentation, suggesting that attention-like switching in insects is tuned to the pace of visual changes in the environment rather than simply the passage of time.

View Article: PubMed Central - PubMed

Affiliation: Queensland Brain Institute, The University of Queensland Brisbane, QLD, Australia.

ABSTRACT
There is increasing evidence that invertebrates such as flies display selective attention (van Swinderen, 2011a), although parallel processing of simultaneous cues remains difficult to demonstrate in such tiny brains. Local field potential (LFP) activity in the fly brain is associated with stimulus selection and suppression (van Swinderen and Greenspan, 2003; Tang and Juusola, 2010), like in other animals such as monkeys (Fries et al., 2001), suggesting that similar processes may be working to control attention in vastly different brains. To investigate selective attention to competing visual cues, I recorded brain activity from behaving flies while applying a method used in human attention studies: competing visual flicker, or frequency tags (Vialatte et al., 2010). Behavioral fixation in a closed-loop flight arena increased the response to visual flicker in the fly brain, and visual salience modulated responses to competing tags arranged in a center-surround pattern. Visual competition dynamics in the fly brain were dependent on the rate of pattern presentation, suggesting that attention-like switching in insects is tuned to the pace of visual changes in the environment rather than simply the passage of time.

No MeSH data available.


Related in: MedlinePlus

Ongoing frequency tag dynamics are tied to pattern velocity. (A) 7 Hz (center) and 9 Hz (surround) power was calculated for a 3 s, moving window following a visual change, for an uninterrupted 40 s (see “Materials and Methods”). (B) Average log-normalized and zero-meaned 7 and 9 Hz power (± SEM) plotted over time, following a novelty event (n = 12 male flies). (C) Average log-ratio of 7 Hz/9 Hz (± SEM) for the same data as in B. *P < 0.05, by t-test compared to zero (dashed line; greater 7 Hz power is above the line, greater 9 Hz below). The number of rotations is indicated below the time axis. (D) The same 7/9 Hz log-ratio analyses were performed on faster (green) and slower (magenta) moving objects. *P < 0.05, **P < 0.01, by t-test compared to zero. The approximate number of rotations for either image velocity condition is indicated by matching color below the time axis. (E) A model of the spatio-temporal dynamics of LFP frequency tag effects following a visual change in the center (red bar). The diameter of the circle indicates the size of a hypothesized attention “spotlight.” The corresponding number of image exposures at different times is indicated for an object moving around the fly at 120°/s. The angle subtended by the different object components (surround, area inside surround, and center) is indicated.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3475995&req=5

Figure 5: Ongoing frequency tag dynamics are tied to pattern velocity. (A) 7 Hz (center) and 9 Hz (surround) power was calculated for a 3 s, moving window following a visual change, for an uninterrupted 40 s (see “Materials and Methods”). (B) Average log-normalized and zero-meaned 7 and 9 Hz power (± SEM) plotted over time, following a novelty event (n = 12 male flies). (C) Average log-ratio of 7 Hz/9 Hz (± SEM) for the same data as in B. *P < 0.05, by t-test compared to zero (dashed line; greater 7 Hz power is above the line, greater 9 Hz below). The number of rotations is indicated below the time axis. (D) The same 7/9 Hz log-ratio analyses were performed on faster (green) and slower (magenta) moving objects. *P < 0.05, **P < 0.01, by t-test compared to zero. The approximate number of rotations for either image velocity condition is indicated by matching color below the time axis. (E) A model of the spatio-temporal dynamics of LFP frequency tag effects following a visual change in the center (red bar). The diameter of the circle indicates the size of a hypothesized attention “spotlight.” The corresponding number of image exposures at different times is indicated for an object moving around the fly at 120°/s. The angle subtended by the different object components (surround, area inside surround, and center) is indicated.

Mentions: In the preceding analyses I contrasted LFP responses to competing flicker before and after a visual change. I next investigated ongoing responses to competing flicker following a visual change. Ongoing 7 and 9 Hz power was calculated and averaged among flies (see “Materials and Methods”) across 40 s of time in a subset of recordings from the 3 s (120°/s) open-loop dataset (Figure 5A). Only recordings lacking a visual change until >40 s were investigated (about 5 of these occurred, by chance, per experiment). These analyses thus queried what happened to 7 and 9 Hz power during the entire 40 s after the last change. Surprisingly, brain responses to the competing tags over 40 s time were highly stereotypical, alternating between favoring the 7 Hz center around 20 s and the 9 Hz surround at 40 s (Figure 5B). A log-ratio analysis of the same data (rather than normalizing each frequency separately, see “Materials and Methods”) shows qualitatively the same result, with significant effects at 20 s in favor of 7 Hz and at 35–40 s in favor of 9 Hz (Figure 5C). Similar tag alternation effects were also seen in experiments on female flies, as well as in males exposed to a different competing frequency, 12 Hz (Figure A2). One interpretation of these results would suggest that when 7 Hz is dominant, flies are attending to the moving center, while suppressing the adjacent surround that is moving with it. Central to this interpretation is the notion that the fly is then actively attending to the flickering object as it moves, and therefore primed to respond at that time to any changes in the center.


Competing visual flicker reveals attention-like rivalry in the fly brain.

van Swinderen B - Front Integr Neurosci (2012)

Ongoing frequency tag dynamics are tied to pattern velocity. (A) 7 Hz (center) and 9 Hz (surround) power was calculated for a 3 s, moving window following a visual change, for an uninterrupted 40 s (see “Materials and Methods”). (B) Average log-normalized and zero-meaned 7 and 9 Hz power (± SEM) plotted over time, following a novelty event (n = 12 male flies). (C) Average log-ratio of 7 Hz/9 Hz (± SEM) for the same data as in B. *P < 0.05, by t-test compared to zero (dashed line; greater 7 Hz power is above the line, greater 9 Hz below). The number of rotations is indicated below the time axis. (D) The same 7/9 Hz log-ratio analyses were performed on faster (green) and slower (magenta) moving objects. *P < 0.05, **P < 0.01, by t-test compared to zero. The approximate number of rotations for either image velocity condition is indicated by matching color below the time axis. (E) A model of the spatio-temporal dynamics of LFP frequency tag effects following a visual change in the center (red bar). The diameter of the circle indicates the size of a hypothesized attention “spotlight.” The corresponding number of image exposures at different times is indicated for an object moving around the fly at 120°/s. The angle subtended by the different object components (surround, area inside surround, and center) is indicated.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3475995&req=5

Figure 5: Ongoing frequency tag dynamics are tied to pattern velocity. (A) 7 Hz (center) and 9 Hz (surround) power was calculated for a 3 s, moving window following a visual change, for an uninterrupted 40 s (see “Materials and Methods”). (B) Average log-normalized and zero-meaned 7 and 9 Hz power (± SEM) plotted over time, following a novelty event (n = 12 male flies). (C) Average log-ratio of 7 Hz/9 Hz (± SEM) for the same data as in B. *P < 0.05, by t-test compared to zero (dashed line; greater 7 Hz power is above the line, greater 9 Hz below). The number of rotations is indicated below the time axis. (D) The same 7/9 Hz log-ratio analyses were performed on faster (green) and slower (magenta) moving objects. *P < 0.05, **P < 0.01, by t-test compared to zero. The approximate number of rotations for either image velocity condition is indicated by matching color below the time axis. (E) A model of the spatio-temporal dynamics of LFP frequency tag effects following a visual change in the center (red bar). The diameter of the circle indicates the size of a hypothesized attention “spotlight.” The corresponding number of image exposures at different times is indicated for an object moving around the fly at 120°/s. The angle subtended by the different object components (surround, area inside surround, and center) is indicated.
Mentions: In the preceding analyses I contrasted LFP responses to competing flicker before and after a visual change. I next investigated ongoing responses to competing flicker following a visual change. Ongoing 7 and 9 Hz power was calculated and averaged among flies (see “Materials and Methods”) across 40 s of time in a subset of recordings from the 3 s (120°/s) open-loop dataset (Figure 5A). Only recordings lacking a visual change until >40 s were investigated (about 5 of these occurred, by chance, per experiment). These analyses thus queried what happened to 7 and 9 Hz power during the entire 40 s after the last change. Surprisingly, brain responses to the competing tags over 40 s time were highly stereotypical, alternating between favoring the 7 Hz center around 20 s and the 9 Hz surround at 40 s (Figure 5B). A log-ratio analysis of the same data (rather than normalizing each frequency separately, see “Materials and Methods”) shows qualitatively the same result, with significant effects at 20 s in favor of 7 Hz and at 35–40 s in favor of 9 Hz (Figure 5C). Similar tag alternation effects were also seen in experiments on female flies, as well as in males exposed to a different competing frequency, 12 Hz (Figure A2). One interpretation of these results would suggest that when 7 Hz is dominant, flies are attending to the moving center, while suppressing the adjacent surround that is moving with it. Central to this interpretation is the notion that the fly is then actively attending to the flickering object as it moves, and therefore primed to respond at that time to any changes in the center.

Bottom Line: There is increasing evidence that invertebrates such as flies display selective attention (van Swinderen, 2011a), although parallel processing of simultaneous cues remains difficult to demonstrate in such tiny brains.Local field potential (LFP) activity in the fly brain is associated with stimulus selection and suppression (van Swinderen and Greenspan, 2003; Tang and Juusola, 2010), like in other animals such as monkeys (Fries et al., 2001), suggesting that similar processes may be working to control attention in vastly different brains.Visual competition dynamics in the fly brain were dependent on the rate of pattern presentation, suggesting that attention-like switching in insects is tuned to the pace of visual changes in the environment rather than simply the passage of time.

View Article: PubMed Central - PubMed

Affiliation: Queensland Brain Institute, The University of Queensland Brisbane, QLD, Australia.

ABSTRACT
There is increasing evidence that invertebrates such as flies display selective attention (van Swinderen, 2011a), although parallel processing of simultaneous cues remains difficult to demonstrate in such tiny brains. Local field potential (LFP) activity in the fly brain is associated with stimulus selection and suppression (van Swinderen and Greenspan, 2003; Tang and Juusola, 2010), like in other animals such as monkeys (Fries et al., 2001), suggesting that similar processes may be working to control attention in vastly different brains. To investigate selective attention to competing visual cues, I recorded brain activity from behaving flies while applying a method used in human attention studies: competing visual flicker, or frequency tags (Vialatte et al., 2010). Behavioral fixation in a closed-loop flight arena increased the response to visual flicker in the fly brain, and visual salience modulated responses to competing tags arranged in a center-surround pattern. Visual competition dynamics in the fly brain were dependent on the rate of pattern presentation, suggesting that attention-like switching in insects is tuned to the pace of visual changes in the environment rather than simply the passage of time.

No MeSH data available.


Related in: MedlinePlus