Some decisions involve more deliberation than others. Even seemingly simple decisions such as those that involve preferences between a pair of familiar items take more time when they involve a choice between options of similar subjective value. This simple observation holds across many kinds of decisions, whether they are based on perception of the environment—is the apple green or red? (Cassey et al., 2013; Gold and Shadlen, 2007; Ratcliff, 2002; Usher and McClelland, 2001)—or on internal values and preferences—do I prefer a green apple or a red one? (Basten et al., 2010; Hunt et al., 2012; Krajbich et al., 2010; Milosavljevic et al., 2010). One explanation for why such decisions take more time is that a commitment to a choice depends on the accumulation of evidence to a threshold, and when the evidence is weaker, more samples are required to reach such a threshold (Krajbich et al., 2010; Milosavljevic et al., 2010). This idea has been studied extensively in perceptual decisions about dynamic stimuli (e.g. moving dots) for which more time clearly provides more samples of external evidence, and therefore can improve the accuracy of the decision (Britten et al., 1996; Britten et al., 1993; Hanks et al., 2015; Mazurek et al., 2003; Newsome and Paré, 1988; Salzman et al., 1990). It is less clear why the same framework would apply to value-based decisions, which depend on internal evidence (Krajbich et al., 2010; Milosavljevic et al., 2010). In such cases, it is not known what the source of the evidence is and why more samples should be required to decide between options that are close in value.

We sought to understand the processes involved in deliberation when making value-based decisions. Our central hypothesis is that the hippocampus plays a key role in this deliberation process, contributing to the comparison between items and the construction of internal samples of evidence bearing on the decision.

This hypothesis is guided by several observations. First, extensive research demonstrates that the hippocampus is necessary for detailed and vivid prospection about future events (Addis and Schacter, 2008; Buckner, 2010; Hassabis et al., 2007; Klein et al., 2002; Race et al., 2011; Schacter et al., 2007). This sort of prospection is likely to guide value-based decisions because it allows a decision-maker to imagine the detailed outcome of each choice option. Second, and more broadly, the hippocampus is known to contribute to relational encoding (Cohen and Eichenbaum, 1993; Horner and Burgess, 2013), a term coined by Cohen and Eichenbaum (1993) to capture the essential role of the hippocampus across many cognitive processes that involve flexible comparison and association between distinct items and features (for reviews, see Barry and Maguire, 2019; Davachi, 2006; Eichenbaum, 2000; Eichenbaum, 2018; Konkel and Cohen, 2009; Palombo et al., 2015a; Shohamy and Turk-Browne, 2013). This relational function of the hippocampus is thought to underlie its well-known role in episodic memory, but the comparison of multiple dimensions of items and their relation to each other is also likely to help guide deliberation during decision making by supplying internal evidence about each option. Recent studies have indeed linked hippocampal-based mnemonic processes to choice behavior by demonstrating that the hippocampus is involved in decisions that explicitly depend on memory by requiring participants to use novel associations acquired in the experiment (Barron et al., 2013; Gluth et al., 2015; Wimmer and Shohamy, 2012). However, a critical open question remains about whether the hippocampus also contributes to seemingly simple decisions—between two highly familiar items—without the explicit demand to use memory.

We conducted two experiments to address this question. First, we conducted an fMRI study in healthy young participants while they made decisions based on well-established subjective value (fMRI; Experiment 1). We reasoned that if the hippocampus supports deliberation, then longer decision times should be related to more engagement of the hippocampus. Second, to test whether the hippocampus plays a causal role in resolving value-based decisions, we tested amnesic patients with damage to the hippocampus and surrounding medial temporal lobe (MTL) as well as age-, education-, and verbal IQ-matched healthy controls (Patients; Experiment 2). Although a choice between two familiar items is not typically thought to depend on the hippocampal memory system (Bartra et al., 2013; Padoa-Schioppa and Assad, 2006; Platt and Plassmann, 2014; Rangel and Clithero, 2014; Rangel et al., 2008), we reasoned that amnesic patients may nonetheless show differences in the way they deliberate about simple value-based decisions. Amnesic patients could take less time because their decisions involve less deliberation, or they could take more time because they try unsuccessfully to deliberate using evidence derived from relational mechanisms. In the latter case, the extra time would not improve their decisions.

In both experiments, participants performed a value-based decision task in which they made a series of choices between two familiar food items (Figure 1). The subjective value of each individual item was determined for each participant using an auction procedure in advance (see Materials and methods), so that we could systematically vary the difference in value between the two items (i.e. ∆Value) during the decision task (see also Grueschow et al., 2015; Krajbich et al., 2010; Milosavljevic et al., 2010; Polanía et al., 2015). The same participants also took part in a control condition in which they made perceptual decisions about the dominant color of a dynamic random dot display (Figure 1 and Figure 1—video 1). The perceptual comparison task solicits the same choice and reaction time behavior but is based on external sensory input.

Figure 1 with 1 supplement see all

Download asset Open asset

In Experiment 1, we found that decision time in the value-based decision task was longer when the choice options were closer in value, as expected (Krajbich et al., 2010; Milosavljevic et al., 2010; Polanía et al., 2015). We also found that reaction times correlated with hippocampal BOLD activity, and this effect was localized to regions of the hippocampus that showed activity related to memory retrieval, independently identified in the same participants. In Experiment 2, we found that amnesic patients were somewhat more stochastic and much slower when making value-based decisions. Importantly, despite parallel behavioral findings in value-based decisions and perceptual decisions in the healthy controls, both the hippocampal BOLD effects and the impairments in patients were selective to the value-based decision task. Together, these findings establish a critical role for the hippocampus in value-based decisions about familiar choice options.

We conducted two experiments to test the mechanisms underlying deliberation in value-based decisions. In the first experiment, we scanned healthy young participants with functional MRI while they performed value-based and perceptual decision tasks. In the second experiment, we tested behavior in amnesic patients with damage to the hippocampus and surrounding MTL as well as age-, education-, and verbal IQ-matched healthy control participants on slightly modified versions of these two decision tasks (see Materials and methods).

Experiment 1: functional MRI

Behavior in both decision tasks conforms to sequential sampling models

On the perceptual decision task, healthy young participants (n = 30) made more accurate decisions when the color was more biased toward blue or yellow (Figure 2A, top) and reaction times (RT) were longer for decisions between options that were more difficult to discriminate (i.e. color coherence near zero, Figure 2A, bottom). Similarly, on the value-based decision task, participants made decisions more consistent with their subjective valuation when ∆Value was larger (Figure 2B, top). RTs were longer for decisions between options for which the magnitude of ∆Value (|∆Value|) was smaller (Figure 2B, bottom). For both the perceptual and the value-based tasks, choices and RT were well described by drift diffusion models (Figure 2, solid lines). This observation is consistent with prior work (Krajbich et al., 2015; Ratcliff and McKoon, 2008; Shadlen and Kiani, 2013) and with the proposal that both types of decisions arise through a process of sequential sampling that stops when the accumulation of evidence satisfies a threshold or bound. The choice functions and range of RT were comparable in the two tasks, as were the goodness of fits (for model parameter estimates, see Figure 2—source data 1; for individual participant fits, see Figure 2—figure supplement 1). Some of the differences between the fits, apparent by eye, are attributed to the different scales of evidence strength in the two tasks (see Figure 2—figure supplement 2). We considered simpler parameterizations of the model, but the full model presented here produced a better fit compared to a model with no power law (BIC = 19.45), and a better fit compared to a model with no power law and flat bounds (BIC = 168.45).

Figure 2 with 3 supplements see all

Download asset Open asset

Figure 2—source code 1

Jupyter notebook with analysis code and output for analyses performed on data from Experiment 1.

https://doi.org/10.7554/eLife.46080.008

Download elife-46080-fig2-code1-v2.ipynb

Figure 2—source data 1

Parameter estimates and goodness of fit measures for Experiment 1.

κ is the drift rate. B₀ is the initial height of the bound. B_del is the delay before the bound starts decreasing. B₂ is the coefficient of the exponential term that governs the bound decrease. t_nd is the non-decision time. σ_tnd is the standard deviation of the non-decision time. μ₀ is the bias in drift rate. Plaw is the coefficient of the power law applied to the stimulus strength (color coherence for perceptual and ∆Value for value-based). NLL is negative log-likelihood of the parameters given the choice and RT data. R² choices is the McFadden pseudo-R² for choice data given color coherence (for perceptual) or ∆Value (for value-based). R² RT is the R² for RT data given color coherence (for perceptual) or ∆Value (for value-based).

https://doi.org/10.7554/eLife.46080.009

Download elife-46080-fig2-data1-v2.xlsx

Figure 2—source data 2

Trial-level data for the perceptual task in Experiment 1.

The file contains seven columns; subject ID, signed color coherence, ${\displaystyle p_{blue}}$, response ('#3'=left, '$4'), reaction time, button order (one means blue requires a left response, two means blue requires a right response), and whether the participant chose blue.

https://doi.org/10.7554/eLife.46080.010

Download elife-46080-fig2-data2-v2.csv

Figure 2—source data 3

Trial-level data for the value-based task in Experiment 1.

The file contains eight columns; subject ID, reaction time, whether the participant chose the item on the right side of the screen, the value placed on the item on the left side of the screen, the value placed on the item on the right side of the screen, the name of the image that appeared on the left, the name of the image that appeared on the right, and the participant’s response ('#3'=left, '$4').

https://doi.org/10.7554/eLife.46080.011

Download elife-46080-fig2-data3-v2.csv

Timing of value-based decisions is related to brain correlates of memory

We first conducted a whole-brain analysis to identify regions in the brain that show (i) an effect of RT: a correlation between RT and BOLD activity for the value-based task more so than for the perceptual task, and (ii) a memory effect: greater BOLD activity for successful retrieval of object memories (using the separate object-memory localizer task, see Materials and methods, Figure 3—figure supplement 1 and Figure 3—source data 1). Each of these analyses of the fMRI data (RT; memory retrieval) identified largely separate networks of brain regions (Figure 3—figure supplement 1 and Figure 3—figure supplement 3; Stark and Squire, 2001; Yarkoni et al., 2009). Critically, however, both showed significant effects in the hippocampus and, as shown in Figure 3 (and Figure 3—source data 2), the conjunction of these two effects revealed significant shared BOLD activity in the hippocampus. BOLD activity in memory-related hippocampal regions was more positively correlated with RT for value-based decisions than perceptual decisions, consistent with our hypothesis that deliberation associated with resolving preference relies on memory-related hippocampal mechanisms.

Figure 3 with 5 supplements see all

Download asset Open asset

Figure 3—source data 1

Activation table for map in Figure 3—figure supplement 1; successful memory retrieval: hits > correct rejections.

https://doi.org/10.7554/eLife.46080.019

Download elife-46080-fig3-data1-v2.xlsx

Figure 3—source data 2

Activation table for map in Figure 3; conjunction between RT effect on BOLD for value-based greater than perceptual with effect of successful memory recognition.

https://doi.org/10.7554/eLife.46080.018

Download elife-46080-fig3-data2-v2.xlsx

Figure 3—source data 3

Activation table for map in Figure 3—figure supplement 2A; overall main effect of value-based greater than perceptual decisions.

https://doi.org/10.7554/eLife.46080.020

Download elife-46080-fig3-data3-v2.xlsx

Figure 3—source data 4

Activation table for map in Figure 3—figure supplement 2B; the effect of RT on BOLD for value-based greater than perceptual decisions, restricted to trials for which the range in RT was matched between the two decision tasks.

https://doi.org/10.7554/eLife.46080.021

Download elife-46080-fig3-data4-v2.xlsx

Figure 3—source data 5

Activation table for map in Figure 3—figure supplement 3A; effect of value-based RT on BOLD.

https://doi.org/10.7554/eLife.46080.022

Download elife-46080-fig3-data5-v2.xlsx

Figure 3—source data 6

Activation table for map in Figure 3—figure supplement 3B; effect of perceptual RT on BOLD.

https://doi.org/10.7554/eLife.46080.023

Download elife-46080-fig3-data6-v2.xlsx

Figure 3—source data 7

Activation table for map in Figure 3—figure supplement 3C; value-based RT > perceptual RT.

https://doi.org/10.7554/eLife.46080.024

Download elife-46080-fig3-data7-v2.xlsx

Figure 3—source data 8

Activation table for maps in Figure 3—figure supplement 3E; Figure 3—figure supplement 3F; Figure 3—figure supplement 3G.

https://doi.org/10.7554/eLife.46080.025

Download elife-46080-fig3-data8-v2.xlsx

Figure 3—source data 9

Activation table for map in Figure 3—figure supplement 5: Modulated effect of the value of the chosen food.

https://doi.org/10.7554/eLife.46080.026

Download elife-46080-fig3-data9-v2.xlsx

We conducted a series of control analyses to consider possible alternative explanations for the differential hippocampal activation on value-based versus perceptual tasks. First, the hippocampal BOLD activity might be related simply to the fact that the value-based decision task makes more demands on memory because it depends on identifying objects. Indeed, a main effect of value-based versus perceptual decisions reveals differences in BOLD activity along the ventral stream and in the medial temporal lobe, including the hippocampus (Figure 3—figure supplement 2A and Figure 3—source data 3). However, if object identification were the reason for the RT effects, one would expect to find only a main effect of task—that is, an overall difference between the two tasks regardless of deliberation time—rather than a significant interaction between task and RT. The observation of both a main effect of task and an interaction with RT suggests that differences in object recognition do not account for the finding in the hippocampus. Second, we wondered whether the hippocampal BOLD activity in the value-based task could be related to the fact that for some participants there was a difference in the range of RT in the value-based task compared to the perceptual task. To test this, we repeated the analysis using only trials that shared the same range of RT on the two tasks (by participant). This analysis revealed a similar result (Figure 3—figure supplement 2B and Figure 3—source data 4), suggesting that the difference in the hippocampus is not related to differences in RT range.

A third possibility we considered was that the tasks differ in overall levels of difficulty. Indeed, RT is a function of the difficulty levels in each of the two tasks, but there is also variability in RT within each level of difficulty, allowing us to address questions about RT while controlling for difficulty. Therefore, we tested the possibility that difficulty accounted for more of the variance in hippocampal BOLD activity than RT by repeating the same analysis as in Figure 3 while controlling for the magnitude of color coherence and ∆Value, as well as other potential correlates of RT (e.g. mean of the pair of values; see Materials and methods). This analysis again revealed RT-related activity in the hippocampus that is greater for value-based than perceptual decisions, even after accounting for other correlates of RT, both within an anatomical ROI of bilateral hippocampus and at a whole-brain corrected level (Figure 3—figure supplement 3 and Figure 3—source datas 5–8). The conjunction between the RT effect and the memory map was again found within the hippocampus ROI (Figure 3—figure supplement 3H). Finally, because our memory encoding task involved value judgments (see Materials and methods), we reran the conjunction analysis using an independent memory recognition localizer that was not specific to value-based encoding, instead using two independent meta-analysis maps from neurosynth.org based on the terms ‘autobiographical memory’ and ‘recollection’. The three-way conjunction between the differential effect of RT on BOLD and these two meta-analysis maps also shows overlap in the hippocampus (Figure 3—figure supplement 4).

Connectivity between hippocampus and parietal cortex increases with value-based decision time

The fMRI results suggest that BOLD activity in the hippocampus is related to the time it takes to make value-based decisions. We next explored the broader neural circuits that interact with the hippocampus during value-based decisions and how activity in such circuits varies with RT. We used a psychophysiological interaction (PPI) analysis to identify brain regions with activity that covaried in an RT-dependent manner with the activity of hippocampal ‘seed’ voxels—that is those that exhibited RT-dependent activation on the value-based decision task and memory-related activation on the memory localizer task. The strongest RT-dependent correlation was between the hippocampus and the parietal cortex (superior parietal lobule and precuneus), showing that functional connectivity between the hippocampus and parietal cortex was greater for value-based decisions that took longer (Figure 4 and Figure 4—source data 1).

Figure 4

Download asset Open asset

Figure 4—source data 1

Activation table for map in Figure 4; PPI for value-based decision trials with hippocampus seed modulated by RT.

https://doi.org/10.7554/eLife.46080.028

Download elife-46080-fig4-data1-v2.xlsx

Experiment 2: behavior in amnesic patients

The fMRI data reveal that the timing of value-based decisions is related to BOLD activity in the hippocampus, suggesting a possible role for the hippocampus in the deliberation process. However, fMRI can only tell us about brain activity correlated with a mental process, leaving open the critical question of whether the hippocampus plays a direct, causal role in value-based decisions. Experiment 2 was designed to address this question by testing value-based decision making in patients with amnesia subsequent to damage to the hippocampus and nearby MTL structures.

Our overarching hypothesis is that the hippocampus contributes to value-based decisions by supporting the comparison of options, the simulation of outcomes, and the recollection of internal evidence. We therefore expected that damage to the hippocampus would impair this deliberation process. As noted earlier, we had no strong prediction regarding whether patients would show faster or slower RTs in general. We reasoned that slower RTs might reflect efforts to search for evidence to resolve decisions, whereas faster RTs might reflect choices that lack deliberative reasoning altogether. Patients with hippocampal damage are not known to have general impairments in valuation processes and the experiment only included food items that each patient fully recognized (see Materials and methods). Therefore, we expected that patients would make choices largely consistent with their subjective valuations. Finally, for the perceptual task, we expected the patients to show intact performance, consistent with the notion that the hippocampus is not needed to make decisions based on external evidence.

Timing of value-based decisions is impaired in amnesic patients

We tested six amnesic patients with damage to the hippocampus and surrounding MTL on the decision tasks from Experiment 1, slightly modified to accommodate the patient population (see Materials ans methods). The patients have well-characterized memory impairments combined with intact verbal reasoning and IQ (see Table 1), and have participated in several prior studies (Foerde et al., 2013; Grilli and Verfaellie, 2016; Palombo et al., 2019; Palombo et al., 2015b). We compared the patients to fourteen age-, education-, and verbal IQ-matched healthy participants.

Table 1

Patient #	Diagnosis	Gender	Age	Edu	WAIS-III		WMS-III			BNT	FAS	L-N sequence	Years since onset
					VIQ	WMI	GM	VD	AD
P01	Hypoxic-ischemic	F	67	12	88	75	52	56	55	−1.3	−1.1	-2	27.29
P02	Status epilepticus + left temp. lobectomy	M	54	16	93	94	49	53	52	−4.6	−0.96	-1	29.17
P03	Hypoxic-ischemic	M	61	14	106	115	59	72	52	0.54	−0.78	1.33	24.18
P04	Hypoxic-ischemic	M	65	17	131	126	86	78	86	1.3	0.03	1.33	15.00
P05	Encephalitis	M	75	13	99	104	49	56	58	−0.11	−0.5	0.33	5.85
P06	Stroke	M	53	20	111	99	60	65	58	1.02	2.1	−0.33	3.45

1. Age in years at first session; Edu, education in years; WAIS-III, Wechsler Adult Intelligence Scale-III (Wechsler, 1997a); WMS-III, Wechsler Memory Scale-III (Wechsler, 1997b); VIQ, verbal IQ; WMI, working memory index; GM, general memory; VD, visual delayed; AD, auditory delayed; scores are age-adjusted such that a score of 100 is the age-adjusted mean with a standard deviation of 15; BNT, Boston Naming Test; FAS, verbal fluency test; L-N, Letter-Number Sequence. BNT, FAS and L-N scores were z-scored against normative data for each test.

On the perceptual decision task, both patients and healthy participants made more accurate decisions when the color was more strongly biased toward blue or yellow (Figure 5A, top). The RTs of both the patients and healthy participants were longer for decisions between options that were more difficult to discriminate (i.e. color coherence near zero, Figure 5A, bottom). Patients took about the same amount of time as healthy controls to make a perceptual decision and there were no significant differences between the groups on accuracy (i.e. slopes of the choice function in Figure 5A, p=0.28) or RT (interaction between |color coherence| and group on RT, p=0.18; and main effect of group on RT, p=0.41). Further, for both groups, choices and RTs were well-described by a drift diffusion model (Figure 5A, solid lines), suggesting that damage to the hippocampus did not impair the patients’ ability to make decisions that require sequential sampling of external evidence.

Figure 5 with 4 supplements see all

Download asset Open asset

Figure 5—source code 1

Jupyter notebook with analysis code and output for analyses performed on data from Experiment 2.

https://doi.org/10.7554/eLife.46080.035

Download elife-46080-fig5-code1-v2.ipynb

Figure 5—source data 1

Parameter estimates and goodness of fit measures for Experiment 2.

κ is the drift rate. B₀ is the initial height of the bound. B_del is the delay before the bound starts decreasing. B₂ is the coefficient of the exponential term that governs the bound decrease. t_nd is the non-decision time. σ_tnd is the standard deviation of the non-decision time. μ₀ is the bias in drift rate. Plaw is the coefficient of the power law applied to the stimulus strength (color coherence for perceptual and ∆Value for value-based). NLL is negative log-likelihood of the parameters given the choice and RT data. R² choices is the McFadden pseudo-R² for choice data given color coherence (for perceptual) or ∆Value (for value-based). R² RT is the R² for RT data given color coherence (for perceptual) or ∆Value (for value-based).

https://doi.org/10.7554/eLife.46080.036

Download elife-46080-fig5-data1-v2.xlsx

Figure 5—source data 2

Trial-level data for the perceptual task in Experiment 2.

The file contains eight columns; subject ID, group (healthy or amnesia), signed color coherence, ${\displaystyle p_{blue}}$, response (‘z’=left, ‘m’), reaction time, button order (one means blue requires a left response, two means blue requires a right response), and whether the participant chose blue.

https://doi.org/10.7554/eLife.46080.037

Download elife-46080-fig5-data2-v2.csv

Figure 5—source data 3

Trial-level data for the value-based task in Experiment 2.

The file contains 12 columns; subject ID, group (healthy or amnesia), the name of the image that appeared on the left side of the screen, the name of the image that appeared on the right, the participant’s response (‘z’=left, ‘m’), reaction time, the value rating of the item on the left, the value rating of the item on the right, whether the participant chose the item on the right side of the screen, the z-scored value rating of the item on the left, the z-scored value rating of the item on the right, and ΔValue.

https://doi.org/10.7554/eLife.46080.038

Download elife-46080-fig5-data3-v2.csv

In contrast, on the value-based decision task the amnesic patients’ performance diverged from that of healthy controls. Although the amnesic patients’ choices were clearly governed by ∆Value (red sigmoid function, Figure 5B top, simple effect of ΔValue on choices among amnesics, p<0.0001), their choices were more stochastic than those of the controls (flatter red sigmoid function, Figure 5B top, p=0.0008). This observation implies that the amnesic patients were not randomly guessing or forgetting the subjective value of the items but were less sensitive to their difference. Notably, the patients did not show any obvious differences in their use of the value rating scale nor in the resulting range of ΔValues (Figure 5—figure supplement 1). This implies that the flatter choice function is not explained by a difference in the use of the value rating scale but that the ∆Value derived from that scale had less purchase on their choices.

The more striking difference between the two groups was observed on RT during value-based decisions: the amnesic patients were substantially slower than healthy controls (Figure 5B bottom, p=0.0004). These slower RTs were specific to the value-based compared to the perceptual decision task (p=0.002 for the interaction between task type and group on RT). In addition, their RTs were less driven by subjective value ratings (flatter red curve in Figure 5B bottom). This difference between amnesic patients and healthy controls was statistically reliable (p=0.015, interaction between |ΔValue| and group on RT in a mixed effects linear regression, see Materials and methods). In principle, slower decisions could be a sign of a speed-accuracy tradeoff favoring accuracy, but that does not appear to be the case, as the patients were both slower and less accurate (i.e. less consistent with initial subjective values) than the controls. To clarify this point, we calculated an index of efficiency (I_E) for each participant (average accuracy divided by the average RT). The index captures the extent to which additional time was used to resolve sources of uncertainty that contribute to stochastic choice behavior. For perceptual decisions, I_E did not differ between amnesic patients and healthy controls (Figure 5C, t_17.21 = 0.02, p=0.98, Welch’s t-test), presumably because the uncertainty originates in the stimulus and its noisy representation by sensory neurons (Britten et al., 1993; Mante et al., 2013; Shadlen and Newsome, 1998). For value-based decisions, I_E was significantly lower in the amnesic patients compared to controls (Figure 5D, t_11.84 = 4.2, p=0.0007, Welch’s t-test). This implies that whatever deliberative process the amnesics engaged in to reach their decisions, it was less efficient than the process used by the controls.

To further characterize differences in the deliberative process between the groups, we evaluated an alternative to the drift-diffusion model. In this ‘heuristic model’, the decision maker makes (1) fast choices for items they like strongly, (2) fast choices for an item paired with one they dislike strongly, and (3) slow stochastic choices when the preference is not resolved by rules 1 and 2 (see Materials and methods and Figure 5—figure supplement 2). The model is representative of a class of alternatives that would account for RT and choice based on distinct rules—that is, a break from sequential sampling with optional stopping. While we found no support for this model in healthy controls (DDM performs better than this heuristic model, BIC = 537.5), at least one feature of the RTs from the amnesic patients is consistent with this model (Figure 5—figure supplement 2). This observation does not provide definitive support for the heuristic above, but it does suggest that the measurable differences between amnesics and controls in accuracy and RT may be related to a fundamental difference in how the amnesics resolve value-based preferences.

We found converging evidence from fMRI and patients pointing to a role for the hippocampus in deliberation between choice options in value-based decisions. In healthy participants, the time it took to resolve choices between two options was longer for near-value decisions and was correlated with BOLD activity in the hippocampus. Amnesic patients with damage to the hippocampus were just as fast as healthy controls to make perceptual decisions but took almost twice as much time to make value-based decisions. The additional time did not lead to better accuracy; in fact, the patients’ choices were less accurate (i.e. more stochastic, relative to the values they initially assigned to the items). Together, these findings link the timing of value-based decisions about highly familiar options to the hippocampus.

Value-based decisions between highly familiar choice options are typically assumed to rely on subjective value (Levy and Glimcher, 2012; Rangel et al., 2008; Tversky and Kahneman, 1986). Such value signals are thought to be supported by the ventromedial prefrontal cortex (vmPFC, Camille et al., 2011; Fellows, 2016; Fellows and Farah, 2007; Levy and Glimcher, 2011; Padoa-Schioppa and Assad, 2006). Yet, even when choosing between options that differ greatly in their subjective value, such choices involve a comparison of the values by way of taking both options, their relation, and their predicted value, into account (Houston et al., 1999; Tversky, 1972; Voigt et al., 2017). Resolving the choice between two options with similar value likely requires the generation of additional information—that is, evidence—to resolve the indecision. This evidence must come from internal sources and might involve multiple dimensions of comparisons between the options. In that sense, it may seem obvious that deliberating between even highly familiar options is likely to involve the sort of relational mechanisms that the hippocampus is known to support.

Our findings suggest that the role of the hippocampus in value-based decisions is almost certainly more nuanced than memory retrieval of the value associated with each of the items. Prior work suggests that simple object-value associations do not depend on the hippocampus (Neubert et al., 2015; Reynolds et al., 2001; Rudebeck et al., 2008; Rushworth et al., 2011; Schultz et al., 1997; Vo et al., 2014). Moreover, it is not obvious why a simple associative memory process would account for longer deliberation times. Instead, we propose that the hippocampus contributes to deliberative processes during decision making. Specifically, we propose that deliberation may be served by the construction of value from internal evidence and engagement in the comparison between the options. Such a process is likely to also involve evaluation of alternatives and prospection about future hypothetical experiences. Prior work suggests that all these processes are likely to engage the hippocampus (Barron et al., 2013; Eichenbaum and Cohen, 2001; Schacter et al., 2007). Future work will be necessary to evaluate how these different processes interact and whether their unique contributions may differ under different circumstances.

Our findings extend recent results demonstrating a role for the hippocampus in value-based decisions under conditions in which value information has been experimentally manipulated to depend on retrieval of new associative memories (Barron et al., 2013; Gluth et al., 2015; Wimmer and Shohamy, 2012). Recent work has also characterized sampling processes during value-based decisions that are reliant on memory (Bornstein and Norman, 2017; Bornstein et al., 2017; Duncan and Shohamy, 2016). Our study builds on these findings to implicate the hippocampus functionally and establish a causal role for the hippocampus in decisions about familiar options for which value is known. One open question is whether this role varies as a function of the nature of the items under deliberation. For example, natural versus packaged items may vary in the extent to which perceptual features reveal their value; the color of an apple is revealing of its sweetness, the color of a package of chocolate perhaps less so. But ultimately, all such decisions depend on the transformation of external perceptual input to internal estimates of subjective value bearing on the relative desirability of the items. It is this deliberative process—beyond the simple item-value association—that we posit the hippocampus contributes to.

The pattern of behavior among the amnesic patients provides further insight into how and when the hippocampus is necessary for value-based decision making. We found that amnesic patients were somewhat less consistent in their decisions and that they took much longer to make them. A similar pattern has been shown recently in healthy older adults with mild memory deficits (Levin et al., 2018). As noted earlier, it is unlikely that amnesic patients simply cannot remember the value of the items, as their choices are not arbitrary. This suggests that the patients may be relying on degraded value signals that are coarser than those in controls. Studies of simple valuation have described general valence signals in neurons in orbitofrontal cortex, striatum, amygdala, and anterior cingulate cortex that could potentially drive these choices (Figure 3—figure supplement 5; Hayden et al., 2009; Hikosaka et al., 2014; Padoa-Schioppa and Assad, 2006; Platt and Plassmann, 2014; Saez et al., 2017). Interestingly, patients with vmPFC damage also show greater stochasticity in their choices (Camille et al., 2011; Fellows and Farah, 2007; Pelletier and Fellows, 2019), but do not display the slowing in RT during deliberation that we see in the patients with amnesia due to hippocampal damage. This finding and others (Jones and Wilson, 2005; Wikenheiser et al., 2017; Wimmer and Büchel, 2016), point to possible complementary roles for the hippocampus and the vmPFC in guiding value-based decisions (also see, McCormick et al., 2018), with the hippocampus possibly supporting evidence-based construction of value and deliberation (Weilbächer and Gluth, 2016).

If patients resolve their choices by accessing a simpler form of value representation, then why do they take such a long time to reach decisions? We propose that this reflects the patients’ attempt to engage hippocampal relational mechanisms and their failure to do so. This conclusion is based on a detailed consideration of the relationship between time, accuracy, and choices. In particular, it may help to elaborate on an important difference between the decision processes at play in the value and perceptual decisions we studied. For both tasks, choice and RT were reconciled by fits to drift diffusion models, indicating that both perceptual and value-based decisions exhibit a systematic relationship in speed and accuracy as a function of difficulty. In the perceptual task, a sequence of samples of blue and yellow dots can be converted by the visual system to samples of evidence by spatially integrating blue or yellow (or the difference) across the stimulus aperture in sampling epochs governed by the temporal resolution of the color system, which is slower than the frame rate of the display. These samples arrive in series until the subject terminates the decision. The samples are independent, identically distributed random values drawn from a distribution with an expectation (i.e. mean) determined by the stimulus strength and a variance governed by the stochastic properties of the stimulus and the neurons that represent blue, yellow or blue minus yellow. The accumulation of these noisy samples is analogous to a deterministic drift plus diffusion.

As mentioned earlier, similar logic has been applied to value-based decisions (Krajbich and Rangel, 2011; Milosavljevic et al., 2010; Polanía et al., 2015), but the analogy breaks down at the nature of the evidence samples. One might posit that neurons that represent value provide the samples of evidence (Rangel et al., 2008; Rangel and Clithero, 2014; Sokol-Hessner et al., 2012). However, the stimulus provides only one sample of the objects, and there is no reason to think that the brain would then generate a sequence of independent samples of ∆Value (Shadlen and Shohamy, 2016). Instead, we reason that the comparison itself triggers constructive thought processes to provide samples of evidence that bear on evaluation of the items along a dimension. It is hard to imagine integrating these samples of ∆Value along different dimensions, although it is possible if they were converted to some common currency (e.g. Kira et al., 2015). It seems at least equally likely that each sample leads to a new internal estimate of preference, only to terminate if such a sample provides a sufficiently compelling preference. Although such a process involves no integration, the drift diffusion model can be fit to such a process well enough to render these alternatives indistinguishable (Ditterich, 2006). On this view, the longer RTs in the amnesic patients stem from their continued effort to generate evidence to resolve the comparison. Accordingly, the greater stochasticity in their choices possibly stems from the fact that they may fail to generate such evidence and ultimately fall back on a more rudimentary and noisier form of value representation to guide their choices. We are not committed to this specific interpretation and consider a simple heuristic strategy that accounts for some aspects of the data (see Figure 5—figure supplement 2).

One limitation of the present study is that we are unable to identify the specific hippocampal-based process that guides deliberation. We can only observe the manifestation of the process in RT and its associated changes in hippocampal BOLD activity or the effect of hippocampal damage. In future work, it will be useful to guide the dimensions of inquiry (e.g. saltiness) and/or construct memories associated with these dimensions that have discernible effects on BOLD activity. In this study, we deliberately avoided any possibility of biasing participants to adopt a memory-based strategy to resolve value preference, as we were interested in testing whether memory spontaneously contributed to such decisions without instruction or guidance.

The idea that memory supports construction of evidence to guide value-based decisions offers new insights to our understanding of how decisions are made, as well as the role of the hippocampus in guiding behavior. The finding that the hippocampus supports deliberation between choice options with similar subjective value addresses a challenge that has long puzzled economists and philosophers (often referred to as Buridan’s paradox, Chislenko, 2016; Sorensen, 2004). By linking the hippocampus to choice behavior, this finding also highlights the pervasive and broad role of the hippocampus in guiding actions and decisions. Research on the hippocampus has typically focused on its role in supporting the formation of conscious, declarative memories for episodes of one’s life. The current findings add to a growing shift in this point of view, suggesting that the hippocampus may serve a more general purpose in guiding behavior by providing behaviorally relevant input about relational associations to implicitly guide actions and decisions (Chun and Phelps, 1999; Eichenbaum and Cohen, 2001; Hannula et al., 2007; Olsen et al., 2016; Palombo et al., 2015a; Ryan et al., 2000; Schapiro et al., 2014; Shohamy and Turk-Browne, 2013; Wimmer and Shohamy, 2012).

Behavioral data from this study are available as source data included in this submission. Behavioral analysis code is available as source code included in this submission. Analysis code as well as task code is available at https://github.com/abakkour/MDMRT_scan (copy archived at https://github.com/elifesciences-publications/MDMRT_scan). Imaging data has been deposited to OpenNeuro and is available to download at https://openneuro.org/datasets/ds002006/versions/1.0.0.

1. 1. Akram Bakkour
   2. Daphna Shohamy
   3. Michael N Shadlen

   (2019) OpenNeuro

   Memory and decision making dataset.

   https://openneuro.org/datasets/ds002006/versions/1.0.0

1. 1. Addis DR
   2. Schacter DL

   (2008) Constructive episodic simulation: temporal distance and detail of past and future events modulate hippocampal engagement

   Hippocampus 18:227–237.

   https://doi.org/10.1002/hipo.20405

   • PubMed
   • Google Scholar
2. Software

   1. Bakkour A

   (2019) MDMRT_scan, version 8e89da1

   GitHub.

   https://github.com/abakkour/MDMRT_scan
3. 1. Barron HC
   2. Dolan RJ
   3. Behrens TE

   (2013) Online evaluation of novel choices by simultaneous representation of multiple memories

   Nature Neuroscience 16:1492–1498.

   https://doi.org/10.1038/nn.3515

   • PubMed
   • Google Scholar
4. 1. Barry DN
   2. Maguire EA

   (2019) Remote memory and the Hippocampus: a constructive critique

   Trends in Cognitive Sciences 23:128–142.

   https://doi.org/10.1016/j.tics.2018.11.005

   • PubMed
   • Google Scholar
5. 1. Bartra O
   2. McGuire JT
   3. Kable JW

   (2013) The valuation system: a coordinate-based meta-analysis of BOLD fMRI experiments examining neural correlates of subjective value

   NeuroImage 76:412–427.

   https://doi.org/10.1016/j.neuroimage.2013.02.063

   • PubMed
   • Google Scholar
6. 1. Basten U
   2. Biele G
   3. Heekeren HR
   4. Fiebach CJ

   (2010) How the brain integrates costs and benefits during decision making

   PNAS 107:21767–21772.

   https://doi.org/10.1073/pnas.0908104107

   • PubMed
   • Google Scholar
7. 1. Becker GM
   2. DeGroot MH
   3. Marschak J

   (1964) Measuring utility by a single-response sequential method

   Behavioral Science 9:226–232.

   https://doi.org/10.1002/bs.3830090304

   • PubMed
   • Google Scholar
8. 1. Beckmann CF
   2. Jenkinson M
   3. Smith SM

   (2003) General multilevel linear modeling for group analysis in FMRI

   NeuroImage 20:1052–1063.

   https://doi.org/10.1016/S1053-8119(03)00435-X

   • PubMed
   • Google Scholar
9. 1. Bornstein AM
   2. Khaw MW
   3. Shohamy D
   4. Daw ND

   (2017) Reminders of past choices Bias decisions for reward in humans

   Nature Communications 8:15958.

   https://doi.org/10.1038/ncomms15958

   • PubMed
   • Google Scholar
10. 1. Bornstein AM
    2. Norman KA

    (2017) Reinstated episodic context guides sampling-based decisions for reward

    Nature Neuroscience 20:997–1003.

    https://doi.org/10.1038/nn.4573

    • PubMed
    • Google Scholar
11. 1. Britten KH
    2. Shadlen MN
    3. Newsome WT
    4. Movshon JA

    (1993) Responses of neurons in macaque MT to stochastic motion signals

    Visual Neuroscience 10:1157–1169.

    https://doi.org/10.1017/S0952523800010269

    • PubMed
    • Google Scholar
12. 1. Britten KH
    2. Newsome WT
    3. Shadlen MN
    4. Celebrini S
    5. Movshon JA

    (1996) A relationship between behavioral choice and the visual responses of neurons in macaque MT

    Visual Neuroscience 13:87–100.

    https://doi.org/10.1017/S095252380000715X

    • PubMed
    • Google Scholar
13. 1. Buckner RL

    (2010) The role of the Hippocampus in prediction and imagination

    Annual Review of Psychology 61:27–48.

    https://doi.org/10.1146/annurev.psych.60.110707.163508

    • PubMed
    • Google Scholar
14. 1. Camille N
    2. Griffiths CA
    3. Vo K
    4. Fellows LK
    5. Kable JW

    (2011) Ventromedial frontal lobe damage disrupts value maximization in humans

    Journal of Neuroscience 31:7527–7532.

    https://doi.org/10.1523/JNEUROSCI.6527-10.2011

    • PubMed
    • Google Scholar
15. 1. Cassey TC
    2. Evens DR
    3. Bogacz R
    4. Marshall JA
    5. Ludwig CJ

    (2013) Adaptive sampling of information in perceptual decision-making

    PLOS ONE 8:e78993.

    https://doi.org/10.1371/journal.pone.0078993

    • PubMed
    • Google Scholar
16. 1. Chislenko E

    (2016) A solution for buridan’s Ass

    Ethics 126:283–310.

    https://doi.org/10.1086/683537

    • Google Scholar
17. 1. Chun MM
    2. Phelps EA

    (1999) Memory deficits for implicit contextual information in amnesic subjects with hippocampal damage

    Nature Neuroscience 2:844–847.

    https://doi.org/10.1038/12222

    • PubMed
    • Google Scholar
18. Book

    1. Cohen NJ
    2. Eichenbaum H

    (1993)

    Memory, Amnesia, and the Hippocampal System

    Cambridge, MA: MIT Press.

    • Google Scholar
19. 1. Davachi L

    (2006) Item, context and relational episodic encoding in humans

    Current Opinion in Neurobiology 16:693–700.

    https://doi.org/10.1016/j.conb.2006.10.012

    • PubMed
    • Google Scholar
20. 1. Ditterich J

    (2006) Stochastic models of decisions about motion direction: behavior and physiology

    Neural Networks 19:981–1012.

    https://doi.org/10.1016/j.neunet.2006.05.042

    • PubMed
    • Google Scholar
21. 1. Drugowitsch J
    2. Moreno-Bote R
    3. Churchland AK
    4. Shadlen MN
    5. Pouget A

    (2012) The cost of accumulating evidence in perceptual decision making

    Journal of Neuroscience 32:3612–3628.

    https://doi.org/10.1523/JNEUROSCI.4010-11.2012

    • PubMed
    • Google Scholar
22. 1. Duncan KD
    2. Shohamy D

    (2016) Memory states influence value-based decisions

    Journal of Experimental Psychology: General 145:1420–1426.

    https://doi.org/10.1037/xge0000231

    • PubMed
    • Google Scholar
23. 1. Eichenbaum H

    (2000) A cortical-hippocampal system for declarative memory

    Nature Reviews Neuroscience 1:41–50.

    https://doi.org/10.1038/35036213

    • PubMed
    • Google Scholar
24. 1. Eichenbaum H

    (2018) What versus where: non-spatial aspects of memory representation by the Hippocampus

    Current Topics in Behavioral Neurosciences 37:101–117.

    https://doi.org/10.1007/7854_2016_450

    • PubMed
    • Google Scholar
25. 1. Eichenbaum H
    2. Cohen NJ

    (2001)

    Memory Systems of the Brain

    From Conditioning to Conscious Recollection, Memory Systems of the Brain, New York, Oxford University Press.

    • Google Scholar
26. Book

    1. Fellows LK

    (2016)

    The Neuroscience of Human Decision-Making Through the Lens of Learning and Memory

    In: Clark R. E, Martin S, editors. Behavioral Neuroscience of Learning and Memory, 37 (2nd ed). Springer International Publishing. pp. 231–251.

    • Google Scholar
27. 1. Fellows LK
    2. Farah MJ

    (2007) The role of ventromedial prefrontal cortex in decision making: judgment under uncertainty or judgment per se?

    Cerebral Cortex 17:2669–2674.

    https://doi.org/10.1093/cercor/bhl176

    • PubMed
    • Google Scholar
28. 1. Fischl B
    2. Sereno MI
    3. Tootell RB
    4. Dale AM

    (1999) High-resolution intersubject averaging and a coordinate system for the cortical surface

    Human Brain Mapping 8:272–284.

    https://doi.org/10.1002/(SICI)1097-0193(1999)8:4<272::AID-HBM10>3.0.CO;2-4

    • PubMed
    • Google Scholar
29. 1. Foerde K
    2. Race E
    3. Verfaellie M
    4. Shohamy D

    (2013) A role for the medial temporal lobe in feedback-driven learning: evidence from amnesia

    Journal of Neuroscience 33:5698–5704.

    https://doi.org/10.1523/JNEUROSCI.5217-12.2013

    • PubMed
    • Google Scholar
30. 1. Gitelman DR
    2. Penny WD
    3. Ashburner J
    4. Friston KJ

    (2003) Modeling regional and psychophysiologic interactions in fMRI: the importance of hemodynamic deconvolution

    NeuroImage 19:200–207.

    https://doi.org/10.1016/S1053-8119(03)00058-2

    • PubMed
    • Google Scholar
31. 1. Gluth S
    2. Sommer T
    3. Rieskamp J
    4. Büchel C

    (2015) Effective connectivity between Hippocampus and ventromedial prefrontal cortex controls preferential choices from memory

    Neuron 86:1078–1090.

    https://doi.org/10.1016/j.neuron.2015.04.023

    • PubMed
    • Google Scholar
32. 1. Gold JI
    2. Shadlen MN

    (2007) The neural basis of decision making

    Annual Review of Neuroscience 30:535–574.

    https://doi.org/10.1146/annurev.neuro.29.051605.113038

    • PubMed
    • Google Scholar
33. 1. Greve DN
    2. Fischl B

    (2009) Accurate and robust brain image alignment using boundary-based registration

    NeuroImage 48:63–72.

    https://doi.org/10.1016/j.neuroimage.2009.06.060

    • PubMed
    • Google Scholar
34. 1. Grilli MD
    2. Verfaellie M

    (2016) Experience-near but not experience-far autobiographical facts depend on the medial temporal lobe for retrieval: evidence from amnesia

    Neuropsychologia 81:180–185.

    https://doi.org/10.1016/j.neuropsychologia.2015.12.023

    • PubMed
    • Google Scholar
35. 1. Grueschow M
    2. Polania R
    3. Hare TA
    4. Ruff CC

    (2015) Automatic versus Choice-Dependent value representations in the human brain

    Neuron 85:874–885.

    https://doi.org/10.1016/j.neuron.2014.12.054

    • PubMed
    • Google Scholar
36. 1. Hanks TD
    2. Kopec CD
    3. Brunton BW
    4. Duan CA
    5. Erlich JC
    6. Brody CD

    (2015) Distinct relationships of parietal and prefrontal cortices to evidence accumulation

    Nature 520:220–223.

    https://doi.org/10.1038/nature14066

    • PubMed
    • Google Scholar
37. 1. Hannula DE
    2. Ryan JD
    3. Tranel D
    4. Cohen NJ

    (2007) Rapid onset relational memory effects are evident in eye movement behavior, but not in hippocampal amnesia

    Journal of Cognitive Neuroscience 19:1690–1705.

    https://doi.org/10.1162/jocn.2007.19.10.1690

    • PubMed
    • Google Scholar
38. 1. Hassabis D
    2. Kumaran D
    3. Vann SD
    4. Maguire EA

    (2007) Patients with hippocampal amnesia cannot imagine new experiences

    PNAS 104:1726–1731.

    https://doi.org/10.1073/pnas.0610561104

    • PubMed
    • Google Scholar
39. 1. Hayden BY
    2. Pearson JM
    3. Platt ML

    (2009) Fictive reward signals in the anterior cingulate cortex

    Science 324:948–950.

    https://doi.org/10.1126/science.1168488

    • PubMed
    • Google Scholar
40. 1. Hikosaka O
    2. Kim HF
    3. Yasuda M
    4. Yamamoto S

    (2014) Basal ganglia circuits for reward value-guided behavior

    Annual Review of Neuroscience 37:289–306.

    https://doi.org/10.1146/annurev-neuro-071013-013924

    • PubMed
    • Google Scholar
41. 1. Horner AJ
    2. Burgess N

    (2013) The associative structure of memory for multi-element events

    Journal of Experimental Psychology: General 142:1370–1383.

    https://doi.org/10.1037/a0033626

    • Google Scholar
42. 1. Houston DA
    2. Doan K
    3. Roskos-Ewoldsen D

    (1999) Negative political advertising and choice conflict

    Journal of Experimental Psychology: Applied 5:3–16.

    https://doi.org/10.1037/1076-898X.5.1.3

    • Google Scholar
43. 1. Hunt LT
    2. Kolling N
    3. Soltani A
    4. Woolrich MW
    5. Rushworth MF
    6. Behrens TE

    (2012) Mechanisms underlying cortical activity during value-guided choice

    Nature Neuroscience 15:470–476.

    https://doi.org/10.1038/nn.3017

    • PubMed
    • Google Scholar
44. 1. Jones MW
    2. Wilson MA

    (2005) Theta rhythms coordinate hippocampal-prefrontal interactions in a spatial memory task

    PLOS Biology 3:e402.

    https://doi.org/10.1371/journal.pbio.0030402

    • PubMed
    • Google Scholar
45. 1. Kira S
    2. Yang T
    3. Shadlen MN

    (2015) A neural implementation of wald's sequential probability ratio test

    Neuron 85:861–873.

    https://doi.org/10.1016/j.neuron.2015.01.007

    • PubMed
    • Google Scholar
46. 1. Klein SB
    2. Loftus J
    3. Kihlstrom JF

    (2002) Memory and temporal experience: the effects of episodic memory loss on an amnesic patient's Ability to Remember the Past and Imagine the Future

    Social Cognition 20:353–379.

    https://doi.org/10.1521/soco.20.5.353.21125

    • Google Scholar
47. 1. Konkel A
    2. Cohen NJ

    (2009) Relational memory and the Hippocampus: representations and methods

    Frontiers in Neuroscience 3:166–174.

    https://doi.org/10.3389/neuro.01.023.2009

    • PubMed
    • Google Scholar
48. 1. Krajbich I
    2. Armel C
    3. Rangel A

    (2010) Visual fixations and the computation and comparison of value in simple choice

    Nature Neuroscience 13:1292–1298.

    https://doi.org/10.1038/nn.2635

    • PubMed
    • Google Scholar
49. 1. Krajbich I
    2. Hare T
    3. Bartling B
    4. Morishima Y
    5. Fehr E

    (2015) A common mechanism underlying food choice and social decisions

    PLOS Computational Biology 11:e1004371.

    https://doi.org/10.1371/journal.pcbi.1004371

    • Google Scholar
50. 1. Krajbich I
    2. Rangel A

    (2011) Multialternative drift-diffusion model predicts the relationship between visual fixations and choice in value-based decisions

    PNAS 108:13852–13857.

    https://doi.org/10.1073/pnas.1101328108

    • PubMed
    • Google Scholar
51. 1. Lagarias JC
    2. Reeds JA
    3. Wright MH
    4. Wright PE

    (1998) Convergence properties of the nelder--mead simplex method in low dimensions

    SIAM Journal on Optimization 9:112–147.

    https://doi.org/10.1137/S1052623496303470

    • Google Scholar
52. 1. Levin F
    2. Fiedler S
    3. Weber B

    (2018) The influence of episodic memory decline on value-based choice

    Aging, Neuropsychology, and Cognition pp. 1–22.

    https://doi.org/10.1080/13825585.2018.1509939

    • Google Scholar
53. 1. Levy DJ
    2. Glimcher PW

    (2011) Comparing apples and oranges: using reward-specific and reward-general subjective value representation in the brain

    Journal of Neuroscience 31:14693–14707.

    https://doi.org/10.1523/JNEUROSCI.2218-11.2011

    • PubMed
    • Google Scholar
54. 1. Levy DJ
    2. Glimcher PW

    (2012) The root of all value: a neural common currency for choice

    Current Opinion in Neurobiology 22:1027–1038.

    https://doi.org/10.1016/j.conb.2012.06.001

    • PubMed
    • Google Scholar
55. 1. Mante V
    2. Sussillo D
    3. Shenoy KV
    4. Newsome WT

    (2013) Context-dependent computation by recurrent dynamics in prefrontal cortex

    Nature 503:78–84.

    https://doi.org/10.1038/nature12742

    • PubMed
    • Google Scholar
56. 1. Mazurek ME
    2. Roitman JD
    3. Ditterich J
    4. Shadlen MN

    (2003) A role for neural integrators in perceptual decision making

    Cerebral Cortex 13:1257–1269.

    https://doi.org/10.1093/cercor/bhg097

    • PubMed
    • Google Scholar
57. 1. McCormick C
    2. Ciaramelli E
    3. De Luca F
    4. Maguire EA

    (2018) Comparing and contrasting the cognitive effects of hippocampal and ventromedial prefrontal cortex damage: a review of human lesion studies

    Neuroscience 374:295–318.

    https://doi.org/10.1016/j.neuroscience.2017.07.066

    • PubMed
    • Google Scholar
58. 1. McLaren DG
    2. Ries ML
    3. Xu G
    4. Johnson SC

    (2012) A generalized form of context-dependent psychophysiological interactions (gPPI): a comparison to standard approaches

    NeuroImage 61:1277–1286.

    https://doi.org/10.1016/j.neuroimage.2012.03.068

    • PubMed
    • Google Scholar
59. 1. Milosavljevic M
    2. Malmaud J
    3. Huth A
    4. Koch C
    5. Rangel A

    (2010) The drift diffusion model can account for the accuracy and reaction time of value-based choices under high and low time pressure

    Judgment and Decision Making 5:437–449.

    https://doi.org/10.2139/ssrn.1901533

    • Google Scholar
60. 1. Neubert FX
    2. Mars RB
    3. Sallet J
    4. Rushworth MF

    (2015) Connectivity reveals relationship of brain Areas for reward-guided learning and decision making in human and monkey frontal cortex

    PNAS 112:E2695–E2704.

    https://doi.org/10.1073/pnas.1410767112

    • PubMed
    • Google Scholar
61. 1. Newsome WT
    2. Paré EB

    (1988) A selective impairment of motion perception following lesions of the middle temporal visual area (MT)

    The Journal of Neuroscience 8:2201–2211.

    https://doi.org/10.1523/JNEUROSCI.08-06-02201.1988

    • PubMed
    • Google Scholar
62. 1. Olsen RK
    2. Sebanayagam V
    3. Lee Y
    4. Moscovitch M
    5. Grady CL
    6. Rosenbaum RS
    7. Ryan JD

    (2016) The relationship between eye movements and subsequent recognition: evidence from individual differences and amnesia

    Cortex 85:182–193.

    https://doi.org/10.1016/j.cortex.2016.10.007

    • PubMed
    • Google Scholar
63. 1. Padoa-Schioppa C
    2. Assad JA

    (2006) Neurons in the orbitofrontal cortex encode economic value

    Nature 441:223–226.

    https://doi.org/10.1038/nature04676

    • PubMed
    • Google Scholar
64. 1. Palombo DJ
    2. Keane MM
    3. Verfaellie M

    (2015a) How does the Hippocampus shape decisions?

    Neurobiology of Learning and Memory 125:93–97.

    https://doi.org/10.1016/j.nlm.2015.08.005

    • PubMed
    • Google Scholar
65. 1. Palombo DJ
    2. Keane MM
    3. Verfaellie M

    (2015b) The medial temporal lobes are critical for reward-based decision making under conditions that promote episodic future thinking

    Hippocampus 25:345–353.

    https://doi.org/10.1002/hipo.22376

    • PubMed
    • Google Scholar
66. 1. Palombo DJ
    2. Di Lascio JM
    3. Howard MW
    4. Verfaellie M

    (2019) Medial temporal lobe amnesia is associated with a deficit in recovering temporal context

    Journal of Cognitive Neuroscience 31:236–248.

    https://doi.org/10.1162/jocn_a_01344

    • PubMed
    • Google Scholar
67. 1. Pelletier G
    2. Fellows LK

    (2019) A critical role for human ventromedial frontal lobe in value comparison of complex objects based on attribute configuration

    The Journal of Neuroscience 39:4124–4132.

    https://doi.org/10.1523/JNEUROSCI.2969-18.2019

    • PubMed
    • Google Scholar
68. 1. Plassmann H
    2. O'Doherty J
    3. Rangel A

    (2007) Orbitofrontal cortex encodes willingness to pay in everyday economic transactions

    Journal of Neuroscience 27:9984–9988.

    https://doi.org/10.1523/JNEUROSCI.2131-07.2007

    • PubMed
    • Google Scholar
69. Book

    1. Platt ML
    2. Plassmann H

    (2014)

    Multistage valuation signals and common neural currencies

    In: Glimcher P. W, Fehr E, editors. Neuroeconomics. Academic Press. pp. 237–258.

    • Google Scholar
70. 1. Polanía R
    2. Moisa M
    3. Opitz A
    4. Grueschow M
    5. Ruff CC

    (2015) The precision of value-based choices depends causally on fronto-parietal phase coupling

    Nature Communications 6:8090.

    https://doi.org/10.1038/ncomms9090

    • PubMed
    • Google Scholar
71. 1. Power JD
    2. Barnes KA
    3. Snyder AZ
    4. Schlaggar BL
    5. Petersen SE

    (2012) Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion

    NeuroImage 59:2142–2154.

    https://doi.org/10.1016/j.neuroimage.2011.10.018

    • PubMed
    • Google Scholar
72. 1. Race E
    2. Keane MM
    3. Verfaellie M

    (2011) Medial temporal lobe damage causes deficits in episodic memory and episodic future thinking not attributable to deficits in narrative construction

    Journal of Neuroscience 31:10262–10269.

    https://doi.org/10.1523/JNEUROSCI.1145-11.2011

    • PubMed
    • Google Scholar
73. 1. Rangel A
    2. Camerer C
    3. Montague PR

    (2008) A framework for studying the neurobiology of value-based decision making

    Nature Reviews Neuroscience 9:545–556.

    https://doi.org/10.1038/nrn2357

    • PubMed
    • Google Scholar
74. Book

    1. Rangel A
    2. Clithero JA

    (2014)

    The computation of stimulus values in simple choice

    In: Glimcher P. W, Fehr E, editors. Neuroeconomics. Academic Press. pp. 125–148.

    • Google Scholar
75. 1. Ratcliff R

    (2002) A diffusion model account of response time and accuracy in a brightness discrimination task: fitting real data and failing to fit fake but plausible data

    Psychonomic Bulletin & Review 9:278–291.

    https://doi.org/10.3758/BF03196283

    • PubMed
    • Google Scholar
76. 1. Ratcliff R
    2. McKoon G

    (2008) The diffusion decision model: theory and data for two-choice decision tasks

    Neural Computation 20:873–922.

    https://doi.org/10.1162/neco.2008.12-06-420

    • PubMed
    • Google Scholar
77. 1. Reynolds JN
    2. Hyland BI
    3. Wickens JR

    (2001) A cellular mechanism of reward-related learning

    Nature 413:67–70.

    https://doi.org/10.1038/35092560

    • PubMed
    • Google Scholar
78. 1. Rudebeck PH
    2. Behrens TE
    3. Kennerley SW
    4. Baxter MG
    5. Buckley MJ
    6. Walton ME
    7. Rushworth MF

    (2008) Frontal cortex subregions play distinct roles in choices between actions and stimuli

    Journal of Neuroscience 28:13775–13785.

    https://doi.org/10.1523/JNEUROSCI.3541-08.2008

    • PubMed
    • Google Scholar
79. 1. Rushworth MF
    2. Noonan MP
    3. Boorman ED
    4. Walton ME
    5. Behrens TE

    (2011) Frontal cortex and reward-guided learning and decision-making

    Neuron 70:1054–1069.

    https://doi.org/10.1016/j.neuron.2011.05.014

    • PubMed
    • Google Scholar
80. 1. Ryan JD
    2. Althoff RR
    3. Whitlow S
    4. Cohen NJ

    (2000) Amnesia is a deficit in relational memory

    Psychological Science 11:454–461.

    https://doi.org/10.1111/1467-9280.00288

    • PubMed
    • Google Scholar
81. 1. Saez RA
    2. Saez A
    3. Paton JJ
    4. Lau B
    5. Salzman CD

    (2017) Distinct roles for the amygdala and orbitofrontal cortex in representing the relative amount of expected reward

    Neuron 95:70–77.

    https://doi.org/10.1016/j.neuron.2017.06.012

    • PubMed
    • Google Scholar
82. 1. Salzman CD
    2. Britten KH
    3. Newsome WT

    (1990) Cortical microstimulation influences perceptual judgements of motion direction

    Nature 346:174–177.

    https://doi.org/10.1038/346174a0

    • PubMed
    • Google Scholar
83. 1. Schacter DL
    2. Addis DR
    3. Buckner RL

    (2007) Remembering the past to imagine the future: the prospective brain

    Nature Reviews Neuroscience 8:657–661.

    https://doi.org/10.1038/nrn2213

    • PubMed
    • Google Scholar
84. 1. Schapiro AC
    2. Gregory E
    3. Landau B
    4. McCloskey M
    5. Turk-Browne NB

    (2014) The necessity of the medial temporal lobe for statistical learning

    Journal of Cognitive Neuroscience 26:1736–1747.

    https://doi.org/10.1162/jocn_a_00578

    • PubMed
    • Google Scholar
85. 1. Schultz W
    2. Dayan P
    3. Montague PR

    (1997) A neural substrate of prediction and reward

    Science 275:1593–1599.

    https://doi.org/10.1126/science.275.5306.1593

    • PubMed
    • Google Scholar
86. 1. Ségonne F
    2. Dale AM
    3. Busa E
    4. Glessner M
    5. Salat D
    6. Hahn HK
    7. Fischl B

    (2004) A hybrid approach to the skull stripping problem in MRI

    NeuroImage 22:1060–1075.

    https://doi.org/10.1016/j.neuroimage.2004.03.032

    • Google Scholar
87. 1. Shadlen MN
    2. Kiani R

    (2013) Decision making as a window on cognition

    Neuron 80:791–806.

    https://doi.org/10.1016/j.neuron.2013.10.047

    • PubMed
    • Google Scholar
88. 1. Shadlen MN
    2. Newsome WT

    (1998) The variable discharge of cortical neurons: implications for connectivity, computation, and information coding

    The Journal of Neuroscience 18:3870–3896.

    https://doi.org/10.1523/JNEUROSCI.18-10-03870.1998

    • PubMed
    • Google Scholar
89. 1. Shadlen MN
    2. Shohamy D

    (2016) Decision making and sequential sampling from memory

    Neuron 90:927–939.

    https://doi.org/10.1016/j.neuron.2016.04.036

    • PubMed
    • Google Scholar
90. 1. Shohamy D
    2. Turk-Browne NB

    (2013) Mechanisms for widespread hippocampal involvement in cognition

    Journal of Experimental Psychology: General 142:1159–1170.

    https://doi.org/10.1037/a0034461

    • PubMed
    • Google Scholar
91. 1. Siegel JS
    2. Power JD
    3. Dubis JW
    4. Vogel AC
    5. Church JA
    6. Schlaggar BL
    7. Petersen SE

    (2014) Statistical improvements in functional magnetic resonance imaging analyses produced by censoring high-motion data points

    Human Brain Mapping 35:1981–1996.

    https://doi.org/10.1002/hbm.22307

    • PubMed
    • Google Scholar
92. 1. Smith SM
    2. Jenkinson M
    3. Woolrich MW
    4. Beckmann CF
    5. Behrens TE
    6. Johansen-Berg H
    7. Bannister PR
    8. De Luca M
    9. Drobnjak I
    10. Flitney DE
    11. Niazy RK
    12. Saunders J
    13. Vickers J
    14. Zhang Y
    15. De Stefano N
    16. Brady JM
    17. Matthews PM

    (2004) Advances in functional and structural MR image analysis and implementation as FSL

    NeuroImage 23 Suppl 1:S208–S219.

    https://doi.org/10.1016/j.neuroimage.2004.07.051

    • PubMed
    • Google Scholar
93. 1. Sokol-Hessner P
    2. Hutcherson C
    3. Hare T
    4. Rangel A

    (2012) Decision value computation in DLPFC and VMPFC adjusts to the available decision time

    European Journal of Neuroscience 35:1065–1074.

    https://doi.org/10.1111/j.1460-9568.2012.08076.x

    • PubMed
    • Google Scholar
94. Book

    1. Sorensen R

    (2004)

    Paradoxes of Rationality

    In: Mele A. R, Rawling P, editors. The Oxford Handbook of Rationality. Oxford University Press. pp. 257–275.

    • Google Scholar
95. 1. Stark CE
    2. Squire LR

    (2001) Simple and associative recognition memory in the hippocampal region

    Learning & Memory 8:190–197.

    https://doi.org/10.1101/lm.40701

    • PubMed
    • Google Scholar
96. 1. Tversky A

    (1972) Elimination by aspects: a theory of choice

    Psychological Review 79:281–299.

    https://doi.org/10.1037/h0032955

    • Google Scholar
97. 1. Tversky A
    2. Kahneman D

    (1986) Rational choice and the framing of decisions

    The Journal of Business 59:S251–S278.

    https://doi.org/10.1086/296365

    • Google Scholar
98. 1. Usher M
    2. McClelland JL

    (2001) The time course of perceptual choice: the leaky, competing accumulator model

    Psychological Review 108:550–592.

    https://doi.org/10.1037/0033-295X.108.3.550

    • PubMed
    • Google Scholar
99. 1. Vo K
    2. Rutledge RB
    3. Chatterjee A
    4. Kable JW

    (2014) Dorsal striatum is necessary for stimulus-value but not action-value learning in humans

    Brain 137:3129–3135.

    https://doi.org/10.1093/brain/awu277

    • PubMed
    • Google Scholar
100. 1. Voigt K
     2. Murawski C
     3. Bode S

     (2017) Endogenous formation of preferences: choices systematically change willingness-to-pay for goods

     Journal of Experimental Psychology: Learning, Memory, and Cognition 43:1872–1882.

     https://doi.org/10.1037/xlm0000415

     • PubMed
     • Google Scholar
101. 1. Wechsler D

     (1997)

     Administration and Scoring Manual

     Wechsler Adult Intelligence Scale, Administration and Scoring Manual, Third Edition (WAIS-III), San Antonio, TX, Psychological Corporation.

     • Google Scholar
102. 1. Wechsler D

     (1997)

     Administration and Scoring Manual

     Wechsler Memory Scale, Administration and Scoring Manual, Third edition (WMS-III), San Antonio, TX, Psychological Corporation.

     • Google Scholar
103. 1. Weilbächer R
     2. Gluth S

     (2016) The interplay of Hippocampus and ventromedial prefrontal cortex in Memory-Based decision making

     Brain Sciences 7:4.

     https://doi.org/10.3390/brainsci7010004

     • Google Scholar
104. 1. Wikenheiser AM
     2. Marrero-Garcia Y
     3. Schoenbaum G

     (2017) Suppression of ventral hippocampal output impairs integrated orbitofrontal encoding of task structure

     Neuron 95:1197–1207.

     https://doi.org/10.1016/j.neuron.2017.08.003

     • PubMed
     • Google Scholar
105. 1. Wimmer GE
     2. Büchel C

     (2016) Reactivation of Reward-Related patterns from single past episodes supports Memory-Based decision making

     Journal of Neuroscience 36:2868–2880.

     https://doi.org/10.1523/JNEUROSCI.3433-15.2016

     • PubMed
     • Google Scholar
106. 1. Wimmer GE
     2. Shohamy D

     (2012) Preference by association: how memory mechanisms in the Hippocampus bias decisions

     Science 338:270–273.

     https://doi.org/10.1126/science.1223252

     • PubMed
     • Google Scholar
107. 1. Yarkoni T
     2. Barch DM
     3. Gray JR
     4. Conturo TE
     5. Braver TS

     (2009) BOLD correlates of trial-by-trial reaction time variability in gray and white matter: a multi-study fMRI analysis

     PLOS ONE 4:e4257.

     https://doi.org/10.1371/journal.pone.0004257

     • PubMed
     • Google Scholar

Author details

1. Akram Bakkour

   Department of Psychology, Columbia University, New York, United States

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   ab4096@columbia.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-6070-4945

2. Daniela J Palombo

   Memory Disorders Research Center, VA Boston Healthcare System and Boston University School of Medicine, Boston, United States

   Present address

   Department of Psychology, The University of British Columbia, Vancouver, Canada

   Contribution

   Supervision, Investigation, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

3. Ariel Zylberberg

   1. Department of Neuroscience, Columbia University, New York, United States
   2. Mortimer B Zuckerman Mind Brain Behavior Institute, Columbia University, New York, United States
   3. Howard Hughes Medical Institute, Chevy Chase, United States

   Contribution

   Conceptualization, Software, Formal analysis, Investigation, Visualization, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2572-4748

4. Yul HR Kang

   Department of Neuroscience, Columbia University, New York, United States

   Present address

   Computational and Biological Learning Lab, Department of Engineering, University of Cambridge, Cambridge, United Kingdom

   Contribution

   Software, Formal analysis, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

5. Allison Reid

   Memory Disorders Research Center, VA Boston Healthcare System and Boston University School of Medicine, Boston, United States

   Contribution

   Data curation, Investigation, Project administration

   Competing interests

   No competing interests declared

6. Mieke Verfaellie

   Memory Disorders Research Center, VA Boston Healthcare System and Boston University School of Medicine, Boston, United States

   Contribution

   Supervision, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

7. Michael N Shadlen

   1. Department of Neuroscience, Columbia University, New York, United States
   2. Mortimer B Zuckerman Mind Brain Behavior Institute, Columbia University, New York, United States
   3. Howard Hughes Medical Institute, Chevy Chase, United States
   4. The Kavli Institute for Brain Science, Columbia University, New York, United States

   Contribution

   Conceptualization, Supervision, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2002-2210

8. Daphna Shohamy

   1. Department of Psychology, Columbia University, New York, United States
   2. Mortimer B Zuckerman Mind Brain Behavior Institute, Columbia University, New York, United States
   3. The Kavli Institute for Brain Science, Columbia University, New York, United States

   Contribution

   Conceptualization, Supervision, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

• 6,717

  views

• 926

  downloads

• 99

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.