Discussion
This study used both self-report and EEG techniques to examine the impact of the number of options on choice overload. The findings showed that increasing the number of choices led to an increased sense of choice difficulty, which in turn caused choice overload. Specifically, choosing from a large range of options was associated with a more negative evaluation of the decision-making process and an increase in avoidance behavior in comparison to a smaller set of options. ERP results revealed that selecting from a larger set of choices resulted in smaller amplitudes of P1 and P2, and a larger amplitude of N2 and LPC, compared to a smaller set. Moreover, MVPA results indicated that there were significant differences in neural activity between large and small choice sets from 128ms to 1500ms. These results will be discussed in more detail below.
ERP results demonstrated the neurological and cognitive processes associated with option evaluation and selection. Compared to the smaller selection set, the amplitudes of P1 and P2 decreased when selecting from the larger selection set. P1 is associated with early visual processing and attention allocation, with its amplitude being enhanced when attention is directed towards a stimulus (Hillyard & Anllo-Vento, 1998; Munneke et al., 2008). P2 is related to early automatic attention allocation and visual processing (Jing et al., 2019). An increased P2 amplitude is observed when a visual stimulus requires a higher level of attentional focus (Handy et al., 2010; Mercado et al., 2006). Previous research has also found that extensive information processing leads to a lower P2 component (Peng et al., 2021). This suggests that choice overload impairs early processing by decreasing the amplitudes of P1 and P2, as individuals must allocate their attentional resources to cognitively process the stimuli and identify their needs while attempting to minimize cognitive effort.
It has been observed that when presented with a small choice set, individuals tend to allocate more attention to the target options in a particular area of the screen. This is in contrast to the large choice set, where fewer cognitive resources are invested. This is due to the fact that complex decisions often require the use of heuristic strategies in order to make decisions quickly (Besedes et al., 2012). Moreover, when faced with a limited number of options, individuals are more likely to employ compensatory strategies in order to thoroughly evaluate the options, thus investing more attentional resources (Besedes et al., 2012; Gerasimou & Papi, 2018). The increased number of choices leads to increased uncertainty and the risk of missing out on the best option. This is reflected in the increased N2 amplitude in the anterior cingulate cortex (ACC) and frontal regions of the brain (Hedgcock et al., 2012; Ma et al., 2010). Additionally, differences in the color, shape, and spatial location of the stimulus can also lead to changes in N2 amplitude (Cui et al., 2000; Tian et al., 2001). Furthermore, research in the field of risky decision-making has revealed that N2 is sensitive to risky information (Wang et al., 2016). Studies related to choice overload have also found that individuals who selected from larger choice sets exhibited cardiovascular responses consistent with high levels of stress (Saltsman et al., 2019). Therefore, it can be concluded that choice overload interferes with early processing and leads to increased cognitive conflict.
It has been suggested that individuals must invest more attentional resources when faced with a larger choice set. However, having more options can also increase the complexity of decision making and reduce an individual’s confidence in their decision-making ability. Research has shown that P3 amplitude is directly correlated with the amount of attentional resources allocated (Folstein et al., 2008) and inversely proportional to the difficulty of decision-making, with a lack of confidence resulting in a smaller P3 (Polich, 1987; Qin & Han, 2009). Furthermore, studies have indicated that lower perceived load triggers larger peaks in P3 amplitude (Barnhardt et al., 2008). This discrepancy may explain the absence of a significant difference in P3.
At the late processing stage, once sufficient information has been gathered, individuals must evaluate the various options and potential outcomes. Late information processing requires more attentional resources due to the lack of initial processing of the large choice set. Thus, individuals assess and analyze options based on their personal needs and the external environment, which leads to increased attentional resource allocation(Zhao et al., 2015). Moreover, individuals tend to experience emotional arousal, such as regret, when selecting from a larger set of options due to counterfactual thinking. This is evidenced by the larger amplitude of the late positive component (LPC) induced by the large choice set compared to the small choice set (Fields, 2023; Hajcak et al., 2006). The LPC amplitude is linked to the allocation of attentional resources and the level of emotional arousal (Hajcak & Foti, 2020; Hajcak & Nieuwenhuis, 2006). Hence, the extensive selection necessitates continuous and prolonged attention, but it also leads to more negative emotional experiences.
Analyses via MVPA revealed that neural activity varied between small and large choice sets from 128ms to 1500ms post-stimulus onset. Temporal generalization analysis illustrated persistent predictive dynamics during the early, middle, and late stages, suggesting prolonged attention engagement. Activation pattern maps indicated that the contrast between the two choice sets depended on activation in the posterior electrode over the anterior electrode during the early stage. In contrast, the late stage displayed an activation pattern opposite to that of the early stage. These findings were partially confirmed by the ERP results, which showed a decrease in the amplitude of the early attention process and an increase in the amplitude of the late attention process in the large choice set condition in comparison to the small choice set condition. This implies the presence of two opposed processes during the early and late stages. Interestingly, P3 amplitude remained unchanged between the two choice set sizes in the middle phase. MVPA successfully distinguished the processing patterns between the choice set sizes with a precision greater than random chance. The activation topography during the middle stage did not demonstrate significant clustering of electrodes, demonstrating that the capacity to differentiate between large and small choice sets during this phase is not reliant on specific electrodes, but instead involves a complex spatial processing pattern. The processing in the middle stage likely involves a complex transition from the early to the late stage.
This study utilized EEG and self-report methods to investigate and refine the cognitive overload theory of choice overload. It was found that choice overload occurs during the information processing phase, not at the time of making the ultimate decision. This finding is consistent with past researches (Lee & Lee, 2004; Reutskaja et al., 2018). Without the pressure of time, individuals take the time to assess all of their options before making a decision, and decision time was not affected by the size of the choice set after a full evaluation of the options. Self-reported data showed that even without time pressure, a significant number of choices still led to negative emotions and an inclination to avoid them. It was further noted that in large choice sets, late sustained attentional processing compensates for the diminished allocation of early attentional resources. Additionally, it was found that people tend to overestimate the effects of choice overload, and there is a lack of consistency between subjective reports of decision difficulty and choice experiences and objective EEG metrics. The findings provided insight into the neurological mechanisms of choice overload, which may help people gain a greater understanding of the choice overload effect and then apply strategies to make better decisions. However, this study had limitations, such as using only landscape pictures as stimulus materials., which are not as complex as the attributes of objects people choose in real-life scenarios. Therefore, future studies should consider using more elaborate options to simulate realistic choice situations.