Enhanced spatial focusing increases feature-based selection in unattended locations

Abstract

Attention is a multifaceted phenomenon, which operates on features (e.g., colour or motion) and over space. A fundamental question is whether the attentional selection of features is confined to the spatially-attended location or operates independently across the entire visual field (global feature-based attention, GFBA). Studies providing evidence for GFBA often employ feature probes presented at spatially unattended locations, which elicit enhanced brain responses when they match a currently-attended target feature. However, the validity of this interpretation relies on consistent spatial focusing onto the target. If the probe were to temporarily attract spatial attention, the reported effects could reflect transient spatial selection processes, rather than GFBA. Here, using magnetoencephalographic recordings (MEG) in humans, we manipulate the strength and consistency of spatial focusing to the target by increasing the target discrimination difficulty (Experiment 1), and by demarcating the upcoming target's location with a placeholder (Experiment 2), to see if GFBA effects are preserved. We observe that motivating stronger spatial focusing to the target did not diminish the effects of GFBA. Instead, aiding spatial pre-focusing with a placeholder enhanced the feature response at unattended locations. Our findings confirm that feature selection effects measured with spatially-unattended probes reflect a true location-independent neural bias.

Figure 1

Experimental design. (a) Experiment 1. Subjects attended to the bicoloured circle in the left VF (dashed line = spatial focus of attention, FOA) and reported by button press whether the half containing the target colour had a convex or concave shape. The simultaneously presented unattended probe (here: red) in the right VF could either be identical to the target colour (match trial, M) or differ from it (non-match trial, NM). Target colour (red, magenta or blue) was assigned blockwise, probe colour (red, magenta or blue) and distractor colour (yellow, green or grey) changed from trial to trial. (b) Experiment 2, placeholder present condition. Subjects attended to the bicoloured circle in the left VF and reported by button press whether its left or right half was drawn in the target colour. Target, probe and distractor colour assignment was as in Experiment 1. To better anchor the subjects’ spatial attention to the location of the upcoming target, the outline of the upcoming bicoloured circle was presented prior to target onset. On half of the trial blocks, the outline presentation was replaced by a blank screen with fixation cross only (placeholder absent condition, not shown here). (c) Derivation of GFBA effects. Effects of global feature-based attention (GFBA) were assessed by comparing brain responses to a colour probe (contralateral activity) as a function of whether it currently matched the attended target colour, or not. The M-NM difference served as an index of GFBA. ERMF waveforms show the averaged signal of influx (blue field lines) and efflux (red field lines, polarity inverted prior averaging) magnetic field maxima.

Figure 2

Behavioural Performance. (a) Experiment 1. The percentage of correct responses and response times are displayed for both the easy and the hard blocks for match (dark grey) and non-match trials (light grey). Subjects were significantly faster and more accurate when performing the easy task. (b) Experiment 2. The percentage of correct responses and response times are displayed for both the placeholder (PH) present and absent condition for match (dark grey) and non-match trials (light grey). While response accuracy was very high with no difference between experimental conditions, the responses were faster in the PH present blocks. In both experiments, responses given on match trials were slightly slower than those given on non-match trials, which is probably caused by an issue of stimulus-response mapping as shown in the Supplementary Data S1. All error bars reflect the standard error of the mean (SEM).

Figure 3

GFBA effects (M-NM difference) of Experiment 1. (a) Overall GFBA modulation sequence (match minus non-match difference, averaged across easy/hard conditions). Magnetic field distribution maps (top view, upper row) and corresponding 3D current source density distribution maps (lateral/back view, lower row) at time points of early and late effect maxima are shown on the right. ERMF waveforms of the early (black trace) and late (grey trace) GFBA modulation on the left display the signal of sensors located at the respective field distribution maxima (sensor sites indicated by black and grey dots in the field distribution maps, signal collapsed over influx (blue field lines) and efflux (red field lines, polarity inverted prior averaging) maximum). (b) Early GFBA modulation. The waveforms show the early GFBA effect of the easy (black solid) and hard (black dashed) condition. Respective field distribution maps and corresponding current source density distributions are displayed on the right. (c) Late GFBA modulation. The waveforms show the late GFBA effect of the easy (grey solid) and hard (grey dashed) condition with corresponding magnetic field maps and current source densities displayed on the right. Sensor sites for the early and late effect (black and grey dots in the field distribution maps) were always chosen at locations of the overall effect maxima (easy/hard average, see (a)). Horizontal black and grey bars (a) as well as black and grey rectangles (b,c) indicate time windows of significant match vs. non-match comparisons as determined for the overall GFBA effects (p < 0.05, corrected for multiple comparisons, as described in Methods).

Figure 4

ERMF response to the target (Experiment 1). (a) Effect of discrimination difficulty on target-related effects. The waveforms show the brain response contralateral to the target on easy (E, black solid) and hard (H, black dashed) trials as well as the hard-minus-easy difference (green solid). The maps on the right show the field distribution and current source density estimate for the hard-minus-easy difference at the respective modulation maximum. The posterior parietal brain response to the discrimination target is significantly enhanced for the high discrimination difficulty targets (hard task) between 225ms-280ms as indicated by the green bar (p < 0.05, corrected for multiple comparisons, as described in Methods). (b) The waveforms show the brain response contralateral to the target (same sensors as in (a)) for match (M, black solid) and non-match (NM, black dashed) trials as well as the match-minus-non-match difference (grey solid). Sliding window t-tests reveal no significant modulation of the target response by a colour match between the target and probe.

Figure 5

GFBA effects (M-NM difference) of Experiment 2. (a) Overall GFBA modulation sequence (match minus non-match difference, averaged across placeholder (PH) absent/present conditions). Magnetic field distribution maps (top view, upper row) and corresponding 3D current source density distribution maps (lateral/back view, lower row) at time points of early and late effect maxima are shown on the right. ERMF waveforms of the early (black trace) and late (grey trace) GFBA modulation on the left display the signal at sensor sites (black and grey dots) chosen at the respective field distribution maxima. Influx (blue field lines) and efflux (red field lines) maxima were averaged after inverting the polarity sign of efflux. (b) Early GFBA modulation. The waveforms show the early GFBA effect of the PH absent (black solid) and PH present (black dashed) conditions. Respective field distribution maps and corresponding current source density distributions are displayed on the right. (c) Late GFBA modulation. The waveforms show the late GFBA effect of the PH absent (grey solid) and PH present (grey dashed) conditions. Corresponding magnetic field maps and current source densities are displayed on the right. Sensor sites for the early and late effect (black and grey dots in the field distribution maps) were always chosen at locations of the overall effect maxima (PH absent/present average, see (a)). Horizontal black and grey bars (a) as well as black and grey rectangles (b,c) indicate time windows of significant match vs. non-match comparisons as determined for the overall GFBA effects (p < 0.05, corrected for multiple comparisons as described in Methods).