An FN400 effect emerged for Price (

An FN400 effect emerged for Price (F(2.70, 911.14) = 5.93, p = .001, ηp2 = .02), which differed slightly by MA group (Figure 2). Follow-up analyses confirmed that buys versus non-buys varied between promotion conditions among High MA. Non-buys (1.55 μV) were less positive than buys (2.47 μV) under no promotions (F(1, 177) = 12.66, p = .000, ηp2 = .07); however, this effect reversed under promotions (F(1, 177) = 12.63, p = .051, ηp2 = .02). Buy amplitudes were less positive (2.65 μV) than non-buys (3.33 μV) among High MA. Among Low MA, buys (2.22 μV) were likewise less positive compared to non-buys (2.91 μV) under promotions (F(1, 159) = 5.45, p = .021, ηp2 = .03), though no effects were observed under no promotions for amplitude. A significant interaction effect of Math Anxiety X PL/C was observed for latency (F(2.85, 960.36) = 6.56, p = .000, ηp2 = .02). Interestingly, this effect was driven by Low MA (F(2.87, 456.24) = 3.82, p = .011, ηp2 = .02). Follow-up analyses localized the effect to within no promotions (F(1, 159) = 7.22, p = .008, η2 = .04). Non-buys (387.63 ms) peaked later than buys (366.39 ms). However, a more selective analysis was unable to find this relationship within 100-millisecond time windows. Thus, it is unclear whether this latency difference indexes differences in peak onset for FN400 or, perhaps more likely, peak onset of an FN400 for buys versus an overlapping P3 for non-buys.