Figure 3. | A cellular mechanism for inverse effectiveness in multisensory integration

Open accessCopyright infoDownload PDF

A cellular mechanism for inverse effectiveness in multisensory integration

Figure 3.

Affiliation details

Brown University, United States; Bard College, Unied States; Roger Williams University, United States
Figure 3.
Download figureOpen in new tabFigure 3. NMDAR activation mediates MSI.

(A) Evoked synaptic responses produced either by a visual stimulus only, a mechanosensory stimulus only, or paired stimuli with a 50 ms interval. Pairing small, subthreshold responses results in large MSI. (B) Average MSIn for suprathreshold (MSIn = 0.12 ± 0.035, n = 10 cells) and subthreshold stimuli (MSIn = 4.34 ± 1.176, n = 6 cells, p=0.0002, Mann-Whitney U = 0). (C) Comparison of arithmetic sum of evoked subthreshold visual and hindbrain responses to evoked crossmodal responses in control and with NMDAR antagonist, APV (50 mM). (D) NMDAR-blocked cells exhibit significant lower levels of MSIn as compared to control cells (Control MSIn = 4.34 ± 1.18, n = 6 cells, NMDAR-block MSIn = 1.11 ± 0.43, n = 10 cells, p=0.016, Mann-Whitney U = 8). (E) Comparison of the linear sum of evoked visual and hindbrain responses (V+H) against the actual evoked crossmodal response (VH). Line indicates linearity. Values above and below diagonal show supralinear and sublinear multisensory responses, respectively. Notice that the APV group more closely approaches linearity. (F) Example loose-cell attached spike recordings to different stimulus conditions with and without APV. Note that NMDAR-blocked cells exhibit stunted supralinear multisensory responses. (G) Comparison of the linear sum of evoked visual and hindbrain spiking responses (V+H) against the actual evoked crossmodal response (VH). (H) Tadpoles in different experimental groups demonstrated different levels of behavioral MSI (ANCOVA F(2,50)=4.1, p=0.02 after adjustment for tadpole responsiveness to unisensory acoustic stimuli: covariate F(1,50)=6.1, p=0.02). In the pharmacological control group, MSI for low-contrast stimuli (0.12 ± 0.13; n = 15) was significantly higher than both zero (one-sample t-test p=0.005), and MSI for high-contrast visual stimuli (−0.10 ± 0.31, n = 23; post-ANCOVA Tukey HSD test p=0.03). The effect of MSI for low-contrast visual stimuli after pharmacological blockade of NMDA receptors with MK801 was close to zero (0.04 ± 0.21, n = 15), and was not significantly different from control MSI for either low- or high-contrast stimuli (Tukey HSD p>0.2).

DOI: http://dx.doi.org/10.7554/eLife.25392.006

Figure 3—source data 1.Source data for Figure 3.

Excel sheet contains individual data points used to generate figure, organized by panel.

DOI: http://dx.doi.org/10.7554/eLife.25392.007

Download source data [figure-3—source-data-1.media-3.xlsx]