The Functions of Consciousness: On the Initiation of Action

Project Awarded: $30,000

Our conscious subjective experience is marked by a vivid internal milieu of sensations, thoughts, and emotions. Despite this rich subjective experience, many aspects of behavior and cognition appear to be able to proceed in an unconscious manner, which raises an important question: what, if any, are the functions of consciousness? In this proposal, we have two primary aims: 1) to investigate the causal role of conscious awareness in action initiation via psychophysical and computational modeling, and 2) to identify the neural substrates of consciousness and their role in action initiation using fMRI. Conventional approaches to studying the functions of consciousness suffer from two primary deficits. First, they conflate differences in consciousness with differences in unconscious perceptual processing (e.g., experiments which compare “seen” and “unseen” trials without considering differences in performance). Second, they also impose a structure on the observer’s perceptual decision-making process which obscures one important role consciousness may play: that of guiding adaptive, voluntary behavior. In this proposal, we have designed a novel behavioral paradigm that explicitly targets the role of consciousness in action initiation while controlling for these two confounds. Specifically, we have designed a psychophysical task which produces matched objective performance across two conditions, but differences in the subject degree of certainty in what has been seen (i.e., conscious experience). Moreover, by incorporating a continuous, free-response version of this psychophysical paradigm, we can explore the degree to which confidence judgements track action initiation in a more ecologically valid environment. We hypothesize that conscious awareness of a stimulus supports voluntary responding to that stimulus. That is, despite matching objective performance capacity between conditions in our psychophysical task, we expect that conditions with greater degrees of subject confidence will lead to more frequent and more accurate voluntary detection behavior. Additionally, we will combine our psychophysical task with state of the art computational and neuroimaging methods to not only quantify how confidence links to voluntary behavior, but identify the neural correlates of how subjective experiences facilitate goal-directed interactions with the external environment. Our novel, interdisciplinary approach holds tremendous promise to yield unique insights about how consciousness subserves our capacity to interact with a dynamic, changing world.

Brian Maniscalco, PhD.  Postdoctoral researcher, Department of Bioengineering, University of California Riverside

Jorge Morales, PhD. Postdoctoral researcher, Psychological and Brain Sciences Department, Johns Hopkins University

Brian Odegaard, PhD. Postdoctoral researcher, Department of Psychology, University of California Los Angeles

The Neural Representation of Counterfactual Thought

Project Awarded: $30,000

Many of the most interesting cognitive feats that humans perform require us to consider not just the things that actually occur, but also non-actual alternative possibilities. Such “counterfactual” thoughts pervade diverse forms of human cognition, from language to morality to causal reasoning, and others. Recent research has identified a set of neural processes involved in explicit, episodic counterfactual thinking; in other words, the deliberative simulation of non-actual events, whether in the past or the future (De Brigard, et al., 2013; De Brigard et al., 2015; Schacter et al., 2015). This work is foundational because it identifies the neural systems that underwrite episodic simulation. Yet, the counterfactual representations that support critical aspects of humans’ ability for language, judgement, and decision-making sometimes occur at an implicit rather than explicit level (Phillips & Cushman, 2017; Phillips & Knobe, 2018).

Our goal is to use fMRI to identify the neural representation of these implicit counterfactual possibilities. Moreover, we propose to adopt a novel approach to multi-voxel pattern analysis that affords stronger inferences about a genuine representational role. Drawing on debates about the nature of representation from philosophy (e.g., Ramsey, 2007; Dretske, 1997), we make a distinction between representational states and ‘mere tracking’ states in order to define empirically discoverable properties of representations. Our analysis protocol is designed to target this distinction and seeks to identify neural representations of counterfactual thought under this more rigorous definition of representation.

We propose to meet these goals through the following Specific Aims: (1) identify neural populations that encode the content of implicit counterfactuals; (2) differentiate the patterns of brain activity associated with implicit counterfactual thought from those associated with thinking about actual events; and (3) establish the representational function of the multivariate patterns identified in Aims 1 and 2.

Should we succeed in achieving our goals, our project will (A) provide the first evidence for how and where implicit counterfactual thought is instantiated in the brain, (B) provide a reliable neural assay of implicit counterfactual representation that can be exploited by researchers in psychology, cognitive science, and philosophy, and (C) demonstrate the ways in which philosophical insights can directly benefit neuroscientific research.

Regan Bernhard, PhD. Postdoctoral researcher in Psychology, Harvard University

Jonathan Phillips, PhD. Postdoctoral researcher in Psychology, Harvard University

Jessey Wright, PhD. Postdoctoral researcher in Psychology, Stanford University 

Semantic Representations in Early Visual Cortex

Project Awarded: $26,500

This project aims to investigate the nature of representations in human early visual cortex. Classic neuroscientific models posit that early visual cortex represents mainly low-level visual features from retinal input. Philosophical accounts by Jesse Prinz and Peter Carruthers posit that consciousness, because it depends on activation of early sensory areas, always has a sensory format (i.e., is a percept or mental image) and is never purely semantic (i.e., a thought). Consequently, there are no conscious thoughts. We aim to challenge both neuroscientific and philosophical accounts by demonstrating that early visual areas can in fact represent semantic information, as a result of feedback from higher areas. We will achieve this by decoding auditory information of several semantic categories from fMRI activity patterns of early visual cortex in the absence of visual stimulation. We will identify which semantic sound categories are represented in early visual cortex and to which level of abstraction.

We propose to examine the semantic/sensory distinction via the abstract/concrete distinction. For example, a semantic representation of a human figure covers a vast amount of possible shapes, orientations, locations, distances, colors, etc., making it abstract. A visualization/perception of a human figure is much more concrete, with a more specific shape, orientation, location, distance, color, etc. we design our experiment in light of this distinction, such that if the decoder performs successful classifications above chance, then it is relying on abstract (hence semantic) information and not on concrete (hence sensory) information.

Demonstrating the existence of semantic representations in early visual cortex, resulting from top-down effects, will challenge the influential neuroscientific and philosophical views described above. Additionally, our results will support a novel account of feedforward vs. feedback effects on early visual cortex. According to it, whereas feedforward information coming from external stimuli or mental imagery is (roughly) concrete, feedback information stemming from high-level areas is at least sometimes abstract. In ordinary visual perception and visualization, both feedback-abstract and feedforward-concrete representations exist in early visual cortex. However, in cases where only sounds are heard, sometimes only the feedback-abstract information is present in early visual cortex.

In sum, the expected empirical results and their philosophical analysis will make room for semantic representations in early sensory areas and for thoughts in consciousness.

Petra Vetter, PhD. Professor, Department of Psychology, Royal Holloway

Assaf Weksler, PhD. Postdoctoral researcher, Department of Philosophy and Psychology, University of Haifa

Workshop: Localizing Representations in the Brain with Neuroimaging Technologies

Project Awarded: $15,000

Neuroscientists investigating human cognition with neuroimaging technology often use ‘representation talk’ to describe their findings. This language is often ambiguous and its empirical and theoretical significance are thus unclear. Sometimes representations are treated as phenomena that are the targets of investigation and intervention, and at other times the concept is used as a stand-in for claims about information processing. To address the consequences of this ambiguity for neuroimaging research we are preparing an interdisciplinary workshop that will critically evaluate the use of neuroimaging technologies as a window into the representational content of neural activity. This workshop is motivated by philosophical work that has argued (e.g., Sullivan 2010) and demonstrated (e.g., Orlandi 2014), that it can be empirically fruitful to treat representations as phenomena to be explained.

Our workshop aims to address three central questions: What is the significance of attributing to a pattern of brain activity the status of ‘representation’? What evidence is required to make such an attribution? Can neuroimaging technologies provide that evidence? We will address the questions by bringing together philosophers whose work directly engages with the concept of representation, and neuroscientists who are concerned with empirical and technical issues that may impede the search for actual representations n the brain. To focus the discussion on practical issues and pressing theoretical challenges, the workshop will be organized around three cognitive domains: action, memory, and perception.

The goal of the workshop is to identify empirical criteria that can identify representational states and their content within the domains of memory, perception and action. Furthermore, by bringing together an interdisciplinary set of researchers, some of whom have collaborated on similar ideas before, our workshop will provide continuing support for the burgeoning interdisciplinary community devoted to addressing theoretical issues in neuroscience through collaboration and conversation between philosophers and neuroscientists. To ensure that the insights obtained from this meeting advance neuroscientific research beyond the practices of those who participate, the workshop will be recorded and the recordings made available online, and the collection of contributions to the workshop will be formed into an edited volume which will be published as a special issue in an open access journal.

Jessey Wright, PhD. Postdoctoral researcher in Psychology, Stanford University

Robyn Waller, PhD. Assistant Professor of Philosophy, Iona University

Sarah Robins, PhD. Associate Professor of Philosophy, University of Kansas

Petra Vetter, PhD. Professor, Department of Psychology, Royal Holloway

The Need for Patterns: Spontaneous Statistical Computation and its Role in Cognition

Project Awarded: $10,800

The human mind does not simply register every event as it happens; instead, it spontaneously summarizes experience by computing summary statistics or patterns. The choice to do so, and under what circumstances, is a design feature of our cognitive system. Here we aim to investigate spontaneous pattern computation and the intrinsic pressures that increase the likelihood of computing them. We take a novel approach by considering pattern computation as a general phenomenon, and leveraging insights from philosophical treatments (Dennett 1987; 1991). Disparate literatures have investigated two forms of patterns: 1) the ability to represent the mean of a set, such as the average size of a set of dots (Alvarez, 2011; Whitney & Yamanashi Leib, 2018) and 2) the ability to detect predictive relations among pairs or triplets of stimuli in space or time (Fiser & Aslin, 2001; Orbán, Fiser, Aslin, & Lengyel, 2008). Aim 1 is to evaluate whether there are common or distinct principles when computing means and contingencies in order to bridge these two literatures. Specifically, a key principle true of means is that they are compressions: the pattern is computed despite losing information about individual items (Alvarez, 2011; Whitney & Yamanashi Leib, 2018). We test whether this principle applies equally to contingency statistics. Aims 2 & 3 are to investigate what kinds of pressures increase our tendency to compute patterns, with a view to understanding the cognitive utility or motivation for pattern computation. We consider two possible (non-exclusive) motivations for computing patterns (Dennett, 1987): one is that we are hungry for patterns: we want to know about any patterns that exist. Another is that we are need-based compressors: we use patterns as an efficient way to deal with limits in our cognitive capacity. Even though compression does take place, a further prediction from the latter view has not been tested: that pattern computation is more likely to happen as it becomes harder to encode individual items. We probe two ways of making item encoding harder: 1) by making sensory input overwhelming by making it very fast; 2) by making items less predictable. We ask whether one or both factors increase pattern computation. If they do, it supports the notion that need-based compression is a motivator of computing patterns. On the other hand, the hungry-for-patterns view predicts that even when patterns are irrelevant, they should be automatically computed. Overall, our goal is to increase our understanding of the relative contribution of different pressures to compute patterns, with the broader aim of revealing core properties of a fundamental cognitive process in which we spontaneously engage every day.

Anna Leshinskaya, PhD. Postdoctoral researcher, Department of Psychology, University of Pennsylvania

Enoch Lambert, PhD. Postdoctoral researcher, Center for Cognitive Studies, Tufts University

Investigating Memory Structure with iEEG

Project Awarded: $7,700

We are exposed to a vast amount of information over the course of a day, and yet our memory system is capable of encoding and later retrieving relevant details about our environment. This would be impossible unless stored information were structured - that is, organized across various dimensions such as space, time and semantic content. Likewise, the most prominent example of memory success, the Method-of-Loci (MoL), can be understood as a way to structure incoming information according to a spatial format. In the MoL, subjects typically memorize a list of unstructured items such as random words or phone numbers by imagining these items in a structured environment such as a familiar childhood walk through the neighborhood. The MoL thus appears to be a way of using a useful feature of spatial memory to store non-spatial content. However, how this process actually works, and why it works so well is poorly understood. Our aim in this study is to investigate how different kinds of structures work together in human memory at the neural level, as a step towards understanding how our memory systems succeed. More specifically, we aim to do by designing a set of memory experiments to be conducted while taking iEEG recordings from neurosurgical patients implanted with electrodes for the treatment of epilepsy. These electrodes reflect the activity from many local neurons, allowing for more precise measurement of localized neural activity than is standardly allowed in investigations of human memory, where invasive recording is not otherwise possible.

The explicit process in the MoL involves two stages: a strategy for encoding items through visualization, and a strategy for retrieving items through a parallel visualization. For example, I would encode a list by imagining a walk through my childhood home, and then later recall the items by imagining walking through the home again and picking up the items. The parallel between encoding and retrieval raises a broader question about memory: how much does the relationship between remembered items at encoding predict subsequent relationships at retrieval? Our task encourages participants to encode lists of items in a spatio-temporal sequence, but allows them to choose freely how to retrieve the items. By varying the semantic cohesion of the lists, we are effectively providing subjects with three possible structuring dimensions: space (a map-like representation of a museum), time (an ordered path through the museum), and semantic content (relations of meaning between items on the list). Based on previous work, we expect for at least spatial structure to be clearly distinguishable at encoding when recording with iEEG from hippocampal areas, and likely temporal structure as well.

The question of how structuring information via mnemonics facilitates remembering is of interest to both philosophers and neuroscientists. Neuroscientists are interested in how the structural features of spatiotemporal encoding and retrieval, as well documented in models of rodent spatial navigation, are conserved and used in human remembering, which makes use of the same neural structures (i.e., the hippocampal formation). The question is also of interest to philosophers as a phenomenon to be explained when theorizing about the nature of memory and exploring relations between levels of explanation in philosophy of science. 

Sara Aronowitz, PhD. Postdoctoral researcher, Department of Psychology, Princeton University

Sarah Robins, PhD. Associate Professor of Philosophy, University of Kansas

Arjen Stolk, PhD. Postdoctoral researcher, Department of Neuroscience, University of California, Berkeley

Stay on Target: Mind Wandering, Prospective Memory, & Responsibility

Project Awarded: $30,450

People lead busy lives, so they need to make plans about what they will do in the immediate and distant future. For example, someone might plan to buy scotch later for the party or pick up her child from daycare on the way home from work. Given that conscious awareness and working memory are limited resources, people cannot focus on carrying out these plans all the time. Instead, people rely on remembering to perform a future task given specific conditions. The representation of this task is stored as a prospective memory. Prospective memories are recalled when a specific cue is detected, which signals the need to activate the relevant features of the intended action. Failures to detect the cue can result in forgetting to complete the prospective task (prospective memory errors). The consequences of prospective memory errors range in severity. For instance, forgetting to buy scotch for a party may bear less severe consequences than forgetting to pick up your child from daycare. People tend to hold others responsible for these errors and the subsequent consequences (cf. Smith 2005; Sher 2009). This suggests that prospective memory errors can be culpable (see Clarke 2014 and Murray 2017 for arguments to this effect that do not rely exclusively on folk judgments of responsibility). Therefore, it is important to establish the conditions under which a prospective memory error is culpable.

One approach to establishing whether a prospective memory error is culpable is considering the circumstances under which the prospective memory error occurs and evaluating whether those circumstances provide a reasonable excuse for the consequences of the prospective memory error. Recent evidence suggests that mind wandering might be one such relevant condition. For instance, mind wandering is a perceptually decoupled state marked by an attentional shift from processing task-relevant stimuli to task-irrelevant thoughts (see Smallwood & Schooler, 2015 for review). Mind wandering is associated with reduced cortical processing of external stimuli (Baird, Smallwood, Lutz, & Schooler, 2014; Kam et al., 2011), which suggests that mind wandering could reduce the likelihood that a prospective memory cue will be detected. This reasoning is further supported by the ‘gateway hypothesis’ of attention that posits attention is directed to either external stimuli or self-generated, internal representations (Burgess, Dumontheil, & Gilbert, 2007). Therefore, prospective memory cues could remain undetected during mind wandering, especially if mind wandering thoughts are unrelated to the prospective task. Conversely, according to the ‘autobiographical planning’ hypothesis, mind wandering thoughts are frequently future oriented, suggesting that mind wandering enables prospective cognitive operations (Baird, Smallwood, & Schooler, 2011). For instance, both spontaneous mind wandering and intentional future-oriented thoughts are associated with activation of the default-mode network—the network of brain regions within the medial surface of the cortex that is active during stimulus-independent, internal thought (Buckner, Andrews Hanna, & Schacter, 2008; Christoff et al., 2009; Preminger, Harmelech, and Malach, 2011; Spreng and Grady 2010)—suggesting convergent neural operations between mind wandering and prospective thinking (Stawarczyk & D’Argembeau, 2015). More generally, evidence suggests that people are more likely to enact plans that they imagine enacting; hence imagery has been used to increase the probability of people voting in elections (Libby, Shaeffer, Eibach, & Slemmer, 2007) and of people successfully acquiring skills (Baumeister & Masicampo, 2011).

Considered in concert, mind wandering might actually facilitate prospective memory cue detection, especially if mind wandering thoughts are related to the prospective task. Lastly, it is possible that the specific prospective memory task and associated cues could modulate the relationship between mind wandering and prospective memory. For instance, according to accounts in visual attention, items associated with subjective assessments of high-value are more likely to capture attention than items associated with low value (e.g., Hickey, Chelazzi, & Theeuwes, 2010). Therefore, mind wandering could impair the detection of prospective memory cues associated with low value prospective memory tasks but not high-value prospective memory tasks. The proposed project will determine the extent to which mind wandering affects prospective memory cue detection by testing these competing hypotheses.

Establishing the relationship between mind wandering and prospective memory is Aim 1 of the current project, and concerns the circumstances under which prospective memory errors occur during periods of mind wandering. Aims 2 and 3 evaluate perceived responsibility for prospective memory performance given mind wandering. Aim 2 of this project will probe the extent to which people hold themselves responsible for their mind wandering and subsequent prospective memory performance after receiving mechanistic and/or intentionalist feedback regarding their mind wandering and the impact of that mind wandering on task performance. We will also measure objective markers of prospective memory and subsequent prospective memory performance in order to test the extent to which feedback affects subsequent performance and self-assessments of responsibility. Aim 3 will then probe how much people hold other actors responsible for 1) the onset and maintenance of mind wandering (Experiment 3a) and 2) the potentially immoral content of mind wandering (Experiment 3b). 

Team members:

Kristina Krasich, PhD student, Department of Psychology, University of Notre Dame

Samuel Murray, PhD student, Department of Philosophy, Universidad de los Andes (Bogotá)

Robert Blakey, PhD student, Departments of Psychology and Criminology, University of Oxford, UK

Epistemic Injustice and Trauma: The Neural Substrates Involved in Having One's Credibility Repeatedly Undermined

Project Awarded: $29,138

Critical inquiry is valuable in healthy public discourse and professional settings. However, it has been suggested that persistent questioning of the capacity of some people to be “knowers” is an important way that racism and sexism are enacted on a societal scale (i.e., testimonial injustice; Fricker, 2007; Washington, 2016). It has been proposed that, for members of marginalized social groups, the day to day experience of repeatedly having one’s credibility undermined, along with other micro-inequities and acts of discrimination, can alter the brain in similar ways to victims of trauma (e.g., reduced volume in parts of the brain associated with affective control like the anterior cingulate and prefrontal cortex, and hyper-responsiveness in regions associated with emotional processing such as the amygdala)--even without one ‘sudden, catastrophic event’ of the sort that qualifies a person for a diagnosis of post-traumatic stress disorder (PTSD) (Carter, 2007; American Psychiatric Association, 2013; for PTSD’s effects on the brain, see Bremner, 2006).

This research investigates whether having one’s testimony persistently and unjustly questioned is experienced behaviorally, physiologically, and neurally in a manner similar to the experience of hostility. If credibility threats can affect the brain in a manner indistinguishable from hostility (hostile threat being typically understood to qualify as a ‘traumatic’ stressor) this would suggest that testimonial injustice is traumatic in a clinically relevant sense. Women, people of color, socioeconomic minorities, and those with prior history of multiple stressors and traumatic life events (i.e., sexual assault and chronic identity-based injustice), are expected to show behavioral, physiological, and neural responses to credibility threats that appear most similar to those responses associated with the experience of hostility, as they may be sensitized to credibility threats as indicative of the presence of other more serious personal threats. Findings consistent with this hypothesis thus lend credence to the idea that testimonial injustice and other ‘microinjustices’ qualify as traumatic for their experiencers, even without the presence of discernibly hostile speech acts.

This research represents novel interdisciplinary work bridging neuroscience and social epistemology within philosophy, and aims to have real social significance. If it turns out that the neural markers of credibility threats are substantially similar (or even indistinguishable) to those associated with PTSD and other forms of traumatic stress (e.g., race-based, sexual assault-based), then we may have good reason to rethink the clinical constructs (e.g., DSM diagnostic criteria and their relationship to cognitive ontology). Moreover, this research directly investigates testimonial injustice as a relatively “soft” way to perpetrate real harm toward targeted groups.

Team members:

Laura Niemi, PhD. Post-doctoral fellow in Philosophy and the Center for Cognitive Neuroscience, Duke University

Clifford Workman, PhD. Post-doctoral fellow in social neuroscience, University of Chicago

Natalia Washington, PhD. Post-doctoral fellow in Philosophy, Neuroscience, and Psychology, Washington University, St. Louis

Why Do Some Stories of Moral Exemplars Increase Prosocial Behavior?

Project Awarded: $28,402.50

We aim to study when and why people are motivated to emulate moral saints, heroes, and other exemplars. Stories about exemplars are often used as sources for moral education, but which stories are most effective at promoting prosocial behavior among students and why?

Previous studies do suggest that the presentation of moral exemplars can induce virtuous behavior via emulation (Bandura, 1969; Kristjánsson, 2006; Sanderse, 2012). Several psychological mechanisms explain why the presentation of moral stories might work—e.g. vicarious social learning (Bandura, 1969; Henrich 2015), moral elevation (Haidt, 2000), and upward social comparison (Blanton, Buunk, Gibbons, & Kuyper, 1999). However, as philosophers have noted, unrealistically high moral standards can also be problematic and even backfire (Wolf, 1982; Carbonell 2012; Curzer, 2015). Psychological studies support this worry, as stories tend to induce more negative responses the more students think the exemplar is irrelevant to their own lives and engages in superhuman deeds that are unattainable (Monin, 2007; Monin et al., 2008).

We have previously conducted lab and classroom experiments that examined how to effectively promote prosocial motivation. The studies measured voluntary service engagement among students in Korea after they read stories of moral exemplars while minimizing potential negative outcomes (Han et al. 2017), such as moral envy and resentment, which were reported in previous experiments (Monin, 2007; Monin et al., 2008). Our studies demonstrated that emulation among students is better promoted by attainable and relevant exemplars, such as peers, compared to unattainable and irrelevant exemplars, such as historical figures. These findings extended previous psychological studies, which have suggested that the attainability and relevance of role models significantly increases emulation of exemplary behavior (Cialdini, 1980; Lockwood & Kunda, 1997).

The following questions remain unresolved. First, previous studies have not clearly illuminated the psychological mechanisms that explain why attainability and relevance of moral exemplars significantly influences emulation. Second, our previous intervention experiments used self-report as the way to measure participants’ prosocial behavior, so they might be susceptible to social desirability bias (Ito & Cacioppo, 2007). Third, we have not tested our interventions among English-speaking participants. Finally, the neurocognitive mechanisms underpinning the efficacy of moral exemplar interventions have not been characterized. In order to address these issues, we plan to employ more sophisticated experimental designs and eventually neuroimaging to clearly investigate the motivational impact of moral exemplars.

Team members:

Hyemin Han, PhD. Assistant Professor of Educational Psychology, University of Alabama

Josh May, PhD. Assistant Professor of Philosophy, University of Alabama at Birmingham

Clifford Workman, PhD. Post-doctoral fellow in social neuroscience, University of Chicago

Sensory Field in Touch

Project Awarded: $22,040

The concept of a visual field is useful both in phenomenological analysis as well as the scientific explanation of visual categorisation and visuomotor guidance. That is, it seems uncontroversial that visual perception ordinarily presents or represents visible stimuli as arrayed in a spatial field within which stimuli are perceived as standing in spatial relations (for discussion see Schwenkler, 2012 and Smythies, 1996). By contrast, although touch is like vision in being a spatial sense, the question whether there is a sensory field in tactile representation is not so easy to answer. In the proposed studies we aim to address this research question: Is it necessary to postulate a tactile field in order to explain the way that information about the spatial features of stimuli is extracted through tactile perception? There are several different ways that spatial information could be extracted through the sense of touch. We are particularly interested in the representation of spatial information in skin space, or the mosaic of receptive fields that are arranged like the cells of a spreadsheet or the pixels of a screen, on the surface of the skin (“skinotopic” space). When an object contacts the skin, it activates a pattern of adjacent receptive fields depending on its shape. Our research question concerns whether tactile stimuli in skin space are represented as arrayed within a spatial field. Thus defined, the concept of a tactile field in skin space is distinct from that of the spatially defined receptive fields of the skin itself: while the latter is a physiological notion, the former is a psychological construct that has its physiological underpinnings on the skin and the relevant parts in the brain. The tactile field defined as such has been studied behaviourally (Serino et al., 2008; Haggard & Giovagnoli, 2011; Fardo et al., in prep). Despite all these results, it remains unclear the extent to which the tactile field computes spatial percepts: In both the 2008 and the 2011 studies, the tactile targets are lines defined by distinct dots, while in the study in preparation the tactile targets are S shapes produced by brushes. The proposed studies will improve on these fronts as well as several others.

Team members:

Antonio Cataldo, PhD. Post-doctoral fellow in cognitive neuroscience, University College London 

Tony Cheng, PhD student, Department of Philosophy, University College London

John Schwenkler, PhD. Assistant Professor, Department of Philosophy, Florida State University

The Perceptual Basis of Confidence

Project Awarded: $20,685

A science of consciousness hinges on our ability to measure subjective perceptual experience. One candidate measure of subjective perception is confidence. When making a perceptual decision—for example, about whether a person in the distance is your friend—observers might be very confident when the decision is easy (e.g. the person is fairly close), but much less confident when the decision is difficult (e.g. the person is quite far). Our goal here is to determine how confidence about the orientation of a visual stimulus is computed in the human brain. Orientation tasks provide a tractable way to study perceptual decision making, as much is already known about how these tasks are represented in the brain. To investigate confidence, we will leverage a recently developed approach to dissociate confidence from accuracy known as “positive evidence bias”. Subjective confidence typically correlates with objective accuracy—an observer will usually identify a nearby person more accurately as well as be more confident about the decision—so dissociating these two types of measures is critical for determining the neural mechanisms that underlie confidence, specifically. In positive evidence bias, confidence differs in two experimental conditions even though accuracy is the same, creating in normal observers a situation mirroring the rare neurological condition known as “blindsight”. In this project, we will combine computational modeling, magnetoencephelography measurements, and behavioral experiments in humans to test mechanisms underlying the computation of confidence. This integrated approach is designed to yield a theoretical account of objective decision and subjective confidence reports across a wide range of orientation judgment tasks, with applicability to perceptual decision-making more broadly. It will, moreover, provide a body of evidence to guide our interpretation of confidence measures, favoring either a perceptual or post-perceptual (i.e. decision-based or metacognitive) basis for confidence. These findings will thus inform our understanding of subjective perception and how best to measure it as well as philosophical debates about the nature of subjective perception, the perception/cognition divide, and the epistemological status of perceptual states and states of confidence.

Team members:

Rachel Denison, PhD. Post-doctoral fellow, Department of Psychology and Neural Science, New York University

Jason Samaha, PhD student, Department of Psychology, University of Wisconsin- Madison

Tony Cheng, PhD student, Department of Philosophy, University College London

The Wandering Brain: Measuring Freely Moving Thought

Project Awarded: $20,160

Philosophers have long been fascinated by the stream of consciousness––thoughts, images, and bits of inner speech that dance hither and thither across the inner stage. Yet for centuries, such “mind-wandering” was deemed essentially private and thus resistant to empirical investigation. We seek to develop an understanding of mind-wandering that is philosophically precise, objectively measurable, and grounded in our knowledge of the brain.

Recent research into streams of thought has focused on a broad, overarching category called task-unrelated thought (TUT), which consists in whatever you think about when you are not focusing on the current task (Smallwood & Schooler, 2006, 2015). Building on our work in philosophy (Irving, 2016a; Irving & Thompson, in press) and theoretical neuroscience (Christoff, Irving, Fox, Spreng, & Andrews-Hanna, 2016), we have recently challenged this view. We have proposed a dynamic framework in which there are several dissociable types of streams of thought broadly and TUT specifically: a wandering subtype that freely moves between topics, as well as two subtypes whose movement is constrained. One constrained subtype is directed towards a specific goal, whereas the other is “stuck” on something emotionally or perceptually salient. We further propose that these subtypes make importantly different contributions to our mental lives, and to agency and wellbeing more broadly. For example, we propose that that the freely wandering subtype of thought specifically makes critical contributions to creativity (Christoff et al., 2016; Sripada, 2016; in press). 

The current project will use novel thought sampling probes, pupillometry, and electroencephalogram (EEG) to dissociate and objectively measure the key subtypes of streams of thought. Thought sampling is a widely-used technique in mind-wandering research, in which participants are asked to rate their preceding thoughts after random interruptions (Smallwood & Schooler, 2006). Our thought sampling questions distinguish the subtypes of thinking in terms of their dynamic features: that is, whether thoughts freely wander or whether they are constrained.

Using these thought sampling questions, we investigate the neural mechanisms that produce distinct subtypes within the stream of thought and build objective markers for these subtypes. Our first experiment will use pupillometry to examine the role of the locus coeruleus norepinephrine (LC-NE) system in producing distinct subtypes of thinking (Franklin, Broadway, Mrazek, Smallwood, & Schooler, 2013; Mittner et al., 2014). We hypothesize that freely wandering thought should be conceptualized as a form of mental exploration. Since the LC-NE system has been implicated in mental exploration (Aston-Jones & Cohen, 2005; Gilzenrat, Nieuwenhuis, Jepma, & Cohen, 2010; Yu & Dayan, 2005), our experiment will therefore test whether the LC-NE system regulates mind wandering (Aston-Jones & Cohen, 2005; Gilzenrat, Nieuwenhuis, Jepma, & Cohen, 2010; Yu & Dayan, 2005). Our second experiment will use EEG to test whether mind-wandering is associated with frontal alpha signatures of creative thinking (Lustenberger, Boyle, Foulser, Mellin, & Fröhlich, 2015), rather than frontal theta signatures of cognitive control (Cavanagh & Frank, 2014). Using the temporally sensitive and multidimensional nature of EEG, we will also attempt to build machine-learning classifiers that could specifically distinguish freely wandering thought from other subtypes. Such a classifier would represent an objective and nonintrusive method for measuring the wandering mind, which could complement or replace subjective thought sampling methods.

Our project integrates philosophical analysis, subjective report, and objective brain measurements. We believe that such an interdisciplinary approach is necessary to quantify what has remained elusive for centuries: the mechanisms that structure the dynamic stream of thought. 

Team members: 

Julia Kam, PhD. Post-doctoral fellow in cognitive neuroscience, University of California, Berkeley

Zachary Irving, PhD. Assistant Professor of Philosophy, University of Virginia

Philosophy, Neuroscience, and Cognitive Ontology

Project Awarded: $6,615

Psychologists study human cognition indirectly by positing constructs such as “episodic memory” and “executive control.” Collectively, these constructs form a “cognitive ontology” (Price and Friston 2005)—a taxonomy of scientifically legitimate psychological kinds. There is now much scientific interest in using neuroscience—particularly functional neuroimaging techniques—to test our cognitive ontology (Poldrack 2010, Lenartowicz et al. 2010, Anderson 2015). For example, Lenartowicz et al. (2010) applied pattern classifiers to fMRI data to test a hypothetical set of cognitive control constructs—e.g., “response inhibition” versus “task switching.” A key assumption of their work, shared by many others (e.g., Price and Friston 2005, Lindquist et al. 2012, Anderson 2014) is that our best psychological theories should align with observed patterns of brain activation.

The idea of using neuroimaging data to revise our cognitive ontology—viz., to discover new cognitive kinds, eliminate existing ones, etc. (Price and Friston 2005, Anderson 2015, Polger and Shapiro 2016) is fraught with empirical and philosophical challenges. One issue is whether informatics efforts—meta--‐analytic databases such as Neurosynth (Yarkoni et al. 2011), the Cognitive Atlas (Poldrack et al. 2011), etc.—can improve inferences from neural to cognitive states (Sullivan 2017). Another is whether analysis techniques such as dimension reduction analyses (Anderson 2014) and multivariate techniques (e.g., multi--‐voxel pattern analysis--‐MVPA) can be used to discover novel cognitive constructs, or to test our cognitive ontology (McCaffrey and Machery 2016, Kaplan and Craver 2016).

The current proposal aims to foster collaboration between philosophers, neuroscientists, cognitive scientists, and computer scientists interested in cognitive ontology. Philosophers studying the mind--‐brain relationship should be aware of developments in the area of cognitive ontology. And, crucially, empirical work on cognitive ontology should be informed by philosophical concerns about the relationship between mechanisms and kinds (Craver 2009, Polger and Shapiro 2016), on the inferential limits of neuroimaging techniques (Ritchie, Kaplan, and Klein 2017), etc. We plan to run a workshop on cognitive ontology at Washington University in St. Louis with and to publish a volume or special journal issue.

Team members:

Joseph McCaffrey, PhD. Post-doctoral fellow, Philosophy-Neuroscience-Psychology Program, Washington University, St. Louis

Testing effects of framing on punishment decisions involving violence risk assessment

Project Awarded: $30,775.50

From stock investments, to medical triage, to criminal sentencing, human beings are regularly tasked to make important decisions with limited information. Unfortunately, the cognitive processes known to enable us to make such decisions efficiently also give rise to systematic biases (Kahneman & Tversky, 1986). Many public institutions have begun to utilize evidence-based protocols to help guard against decision bias, yet little is known about the way practitioners carry out and interpret the results of these protocols. Nowhere is this problem more concerning than in courts of law, where authorities must make punishment decisions based on tools designed to estimate violence risk. The overall aim of this project is to characterize the influence of two potential sources of bias on punishment judgments in the context of how violence risk assessments are communicated to judges and jurors: (1) whether violence risk estimates are based on evidence that is framed as behavioral or neurobiological, and (2) whether these estimates are expressed using a loss frame (e.g., “26% probability of violence recurring”), gain frame (e.g., “74% probability of violence not recurring”), or both frames. This aim will be achieved by conducting three experimental vignette surveys with 500 jury-eligible U.S. citizens and 200 professional judges. Using case summaries adapted from real criminal cases, these studies will test hypotheses that framing negatively impacts punishment judgments among lay and expert judges, and that particular personality traits or competencies help to explain when a person will be most susceptible to such framing.

Team members:

Eyal Aharoni, Ph.D. Psychology Georgia State University, Atlanta, GA

Jennifer Blumenthal-Barby, Ph.D. Baylor College of Medicine Houston, TX 

Gidon Felsen, Ph.D. University of Colorado School of Medicine, Aurora, CO

Karina Vold, Ph.D/Candidate Department of Philosophy, McGill University, Canada

Flexible Interaction as a Criterion for Consciousness

Project Awarded: $29,820

The study of consciousness is one of the biggest challenges facing modern science. In perceptual psychology report-based measures are the ‘gold-standard’ for attributing ‘consciousness of an object’ (henceforth consciousness) to a subject. There are, however, long-standing problems with the reliability of report as a measure of consciousness. Other accounts have attempted to avoid these problems by identifying objective measures that are correlated with consciousness. However, these ‘no-report’ paradigms continue to rely on report for validation of their measures. Our goal is to identify an objective measure of consciousness that does not rely on report for validation and cannot be attributed to adaptive unconscious processes. Here, we propose one such measure and seek to test whether or not it is capable of dissociating consciousness, reportability, and unconscious, automatic processes in normally-sighted populations and in patients who possess visual field deficits following brain injury.

In order to develop an objective measure of consciousness, we will appeal to intentional accounts of consciousness that claim a subject is conscious of an object if information about that object is available to them for use in explicit reasoning and intentional action. We will argue that reasoning and intentional action are dissociable, and that evidence of intentional action is sufficient for the attribution of consciousness to a subject. Finally, we will argue that intentional action can be differentiated from automatic, unconsciously mediated behavior by virtue of the flexible use of information in performing a task. This line of argument will lead us to posit a ‘flexible interaction criterion for consciousness’ (FI) that says: If a subject exhibits the capacity to use information about an object to reliably guide their action in an object-appropriate and flexible fashion, they are conscious of that object.

In the empirical phase of this project we will test two competing hypothesis. Hypothesis (1) states that the process that makes information available for reasoning and intentional action is an all or nothing process. Hypothesis (2) states that different processes make information available for use in different tasks. As such, (1) and (2) make different predictions regarding the dissociability of FI from reportability with (1) predicting that they will not dissociate and (2) predicting that they will. Our intention is to investigate these two possibilities in both normal subjects and in subjects with selective brain injury.

Team members:

Robert Foley, Rotman Postdoctoral Fellow in the Philosophy of Neuroscience, The Rotman Institute of Philosophy and The Brain and Mind Institute, The University of Western Ontario

Robert Whitwell, Postdoctoral Fellow in the Department of Psychology, The University of British Columbia

The Mental Structure of Moral Judgments. Causality, Intentionality, and Responsibility

Project Awarded: $28,350

Attributing moral responsibility for actions involves establishing the causal relationship between events. However, it also involves making inferences about people’s mental states, including their intentions, desires, and knowledge about the world. This is reflected both in folk psychological conceptions of moral responsibility, and among some normative ethical theories, such as intent-based versions of utilitarianism (Adams 1976). Inferences about individual mental states have been shown to be the most relevant factor when people judge an act to be wrong or permissible, but assigning blame or responsibility for the outcome also depends on a perceived causal link between the agent and the outcome (Cushman 2008). Intentionality judgments also rely on the valence and value of an action’s outcome, the objective probability of an action’s having a desired effect, and the agent’s subjective probability of the action having its effect (Knobe 2003; Feltz 2007; Mele & Cushman 2007). Causal judgments thus clearly affect attributions of moral responsibility. However, it remains unclear how sensitive people are to differences in the causal structure of events with morally relevant content and how intentionality ascriptions emerge from observations (or descriptions) of behaviour, in particular when explicit information about an agent’s intentions is not available. It further remains to be seen to which features of a set of events people preferentially attend when making moral judgments. To clarify these important issues, we propose a series of experiments that will all bear on one overarching question: “What is the mental structure of responsibility attributions?” More specifically, we intend to answer two questions (with some sub-questions) in the realm of descriptive moral psychology:

1.     How sensitive are people to differences in the causal structure of a given causal chain in making cued judgments of intentionality and responsibility?

a.     How dissociable are intentionality and responsibility judgments under conditions of different background causal chains in the absence of explicit intentionality information?

b.     Do causal structure differences between action and non-action (omission) engage differential activation of motor representation regions and does this activation inform responsibility judgment behavior directly, or is this influence mediated by intentionality computations?

c.     What role do counterfactual representations of the contrastive set of unrealized events play in facilitating intentionality responsibility judgments?

2.     What is the latent structure of event/agent features to which people attend when evaluating actions without being explicitly cued to respond to one feature or another?

Experimentally, these questions will be answered by asking study participants to react to vignettes involving interactions between agents that result in morally relevant outcomes. The first set of experiments will involve using vignettes with maximally similar cover stories while varying the causal structure of the action (i.e. action or omission and preemptive or direct causality). The second set of experiments will involve using vignettes based on real episodic memories from prior research participants. In summary, the first experimental set is expected to clarify to what extent intention and responsibility judgments are subject to influence by the background structure of causal relations in the world. The second experimental set will further clarify the latent structure underlying folk psychological conceptions of moral responsibility by using data reduction techniques.

Team members:

Dr. Sofia Bonicalzi

Dr. Chiara Brozzo

Dr. Eugenia Kulakova

Joachim Operskalski

Morality Under Uncertainty: The impact of moral framing on spatial decision making

Project Award: $17,314.50

Recent research on the topic of decision-making suggests that participants make suboptimal decisions involving spatial cues in a risky environment. Participants appear to neither maximize gain nor minimize risk in a task in which they are told to identify the locus of a target that is located in close proximity of a penalty. Given that previous research has suggested that the moral or social frame of decisions can result in participants having an alternative judgment in said task, and that there is evidence that participants behave differently in morally salient perceptual tasks when compared to controls, we propose a set of studies to investigate if the moral framing of perceptual decision tasks results in a difference in the character of the decisions participants make. First, our project involves the use of moral and nonmoral scenarios in risky decision-making tasks, to determine the degree to which participants choose in a more or less optimal manner in a morally characterized task. With this, we can determine if any decision frame has an effect when compared to previous research, and if the moral decision frame has a unique effect. Second, we intend to test for the moral disposition of the participant, in order to determine if the effect and the direction of the moral framing correlates with the degree with which the participant is dispositionally a utilitarian or deontologist. This allows us to determine if moral disposition is predictive of decision-making in spatial tasks, and if differences in disposition track trends in interpretation of success in a spatial task that involves both risk and reward. Third, we intend to pair the behavioral research with neuroimaging, in order to determine what neural network is implicated by activity in the task, and compare it to a previously identified network putatively implicated in decision-making. If we uncover systematic behavioral differences, we wish to investigate the relationships between the networks that underlie the actions represented by the respective behavioral trends. This will elucidate the neural correlates of moral decision-making by implementing a type of stimulus that is under-investigated in moral psychology. Moreover, it helps clarify the degree to which neural networks are conserved across tasks that are prima facie similar to one another, by attempting to elicit dissimilar behaviors from participants. Finally, it will facilitate a better understanding of the circumstances under which decision-making behaviors vary, and the degree to which the character of situation surrounding the decision results in alternative behavior trends.

Team members:

David Colaco, PhD Student,History of Philosophy and Science, University of Pittsburgh and Center for the Neural Basis of Cognition

Kevin Jarbo, PhD Student Psychology, Carnegie Mellon

Beyond the essential moral self: the importance of social relationships in judgments of first- and third­person identity change

Project Award: $30,240

Clarifying the nature of the self and identity has wide ranging implications for a number of issues in philosophy, psychology, and neuroscience. This project aims to better understand the nature of self and identity by examining judgments of identity change and the mechanisms that underlie them. Well worn arguments from the history of philosophy suggest that memory is at the core of personal identity, such that switching out memories is akin to switching out identities (Locke 1690/2009). Recent empirical work, however, suggests that morality (Strohminger & Nichols 2014, 2015), and more specifically, social relationships (Heiphetz et al. 2016), are at the core of personal identity, but the underlying reasons remain unclear.

Our project aims to replicate and refine this Moral Self Effect and, for the first time, examine the neural underpinnings of the essential moral self. Specifically, we investigate whether moral traits may be tied to values, in which case we would expect the effect to hold more strongly for domains that people value (e.g., self; community). In our behavioral studies, we will first examine whether there are asymmetries between first and third person judgments of identity change that is, whether people think differently about changes in themselves vs. others. Drawing on intergroup social psychological research, we will then further refine the Moral Self Effect by: 1) examining whether the valence of moral traits affects judgments of identity change; 2) examining how changes to social relationships affect judgments of identity change; and 3) examining whether the dimensions of warmth and competence might better explain perceived identity change. What will emerge from this suite of behavioral studies is a more refined answer to the question: which parts of the self and community are perceived as the most essential?

Our behavioral results, in turn, will drive our innovative theoretical approach to questions of how perceptions of identity change are represented in the brain, which will help to clarify reasons why certain aspects of self and community might be perceived as most essential. A traditional view in the neuroscience literature is that cortical midline structures track the degree of similarity to the self in a ventral to dorsal fashion (Mitchell, Macrae, & Banaji, 2006), but more recently it has been suggested that these structures track personal value rather than self reference (Kim & Johnson, 2015). We will employ the fine grained and philosophically informed research on the moral self effect in an fMRI task to shed new light on this debate and determine whether distinct neural architectures support first vs. third person judgments of identity change or whether they track something deeper, such as value. Together, this suite of experiments will refine and extend the moral self theory to explain those aspects of identity that are perceived to be most essential.

Team members:

Jim A.C. Everett, D.Phil Candidate, University of Oxford

Michael Ferguson, Postdoctoral Fellow, Cornell University, Human Development

Jordan L. Livingston, PhD Candidate, University of Oregon, Department of Psychology

Joshua August Skorburg, PhD Candidate, University of Oregon, Department of Philosophy

Testing the Predictive Coding Theory of Attention

Project Awarded: $21,000

On the predictive coding (PC) view the mind engages in Bayesian hypothesis generation, testing and revision with the aim of minimizing the mismatch between its predictions and the sensory evidence (prediction error). Though the bulk of the empirical support for the theory lies in the perceptual domain (e.g. Hohwy, et al. 2008; Huang & Rao, 2011; Stefanics et al. 2014), the

Bayesian framework also promises to deliver a comprehensive theory of attention that falls out of the tools employed in the perceptual theory, without the need for positing additional machinery.

According to this proposal, in all instances of attentional behavior the selectivity of attention is explained on the basis of expected precision; attention selects a signal for further processing because the brain expects that signal to deliver a more precise or consistent set of data. The PC theory of attention is thus committed to the claim that high precision expectations are driving attention in all its instances.

However, there has been negligible empirical investigation of this claim to date (Egner, Monti & Summerfield 2010; Jiang, Summerfield & Egner 2013; Kok et al. 2011). We propose to address this gap by using affect-biased attention (ABA) as a test case for the theory, because it is a well studied phenomenon that at least prima facie poses a challenge to PC’s claim. ABA is attention to stimuli that are affectively salient, i.e. stimuli that stand out because the agent associates them with reward or punishment (Todd, Cunningham, Anderson, & Thompson, 2012). ABA has properties of both bottom-up and top-down attention biasing, and so is not easily captured by the PC model. Affectively salient objects can capture attention even when they are not physically salient (Niu et al. 2012a, 2012b; Awh et al. 2012), preventing straightforward assimilation to predictive coding’s treatment of bottom up attention. In addition, affectively salient objects can capture attention when they are not task relevant (Awh et al. 2012; Todd et al. 2012), preventing straightforward assimilation to PC’s treatment of top down attention. To assess the explanatory adequacy of the PC theory of attention, therefore, we propose to tease apart precision expectation and affective salience experimentally, probing their respective effects on the behavioral responses evoked by visual stimuli. By manipulating these two variables independently, it should become clear whether or not PC can accommodate affective biases on attention.

Team members:

Sina Fazelpour, UBC, PhD student, Department of Philosophy

Madeleine Ransom, UBC, PhD student, Department of Philosophy

Jelena Markovic, UBC, PhD student, Department of Philosophy