Jonathan Scholz

Distinct regions of right temporo-parietal junction are selective for theory of mind and exogenous attention.

In functional magnetic resonance imaging (fMRI) studies, a cortical region in the right temporo-parietal junction (RTPJ) is recruited when participants read stories about peoples thoughts (‘Theory of Mind’). Both fMRI and lesion studies suggest that a region near the RTPJ is associated with attentional reorienting in response to an unexpected stimulus. Do Theory of Mind and attentional reorienting recruit a single population of neurons, or are there two neighboring but distinct neural populations in the RTPJ? One recent study compared these activations, and found evidence consistent with a single common region. However, the apparent overlap may have been due to the low resolution of the previous technique. We tested this hypothesis using a high-resolution protocol, within-subjects analyses, and more powerful statistical methods. Strict conjunction analyses revealed that the area of overlap was small and on the periphery of each activation. In addition, a bootstrap analysis identified a reliable 6–10 mm spatial displacement between the peak activations of the two tasks; the same magnitude and direction of displacement was observed in within-subjects comparisons. In all, these results suggest that there are neighboring but distinct regions within the RTPJ implicated in Theory of Mind and orienting attention.

Brain Regions for Perceiving and Reasoning About Other People in School‐Aged Children

Neuroimaging studies with adults have identified cortical regions recruited when people think about other peoples thoughts (theory of mind): temporo-parietal junction, posterior cingulate, and medial prefrontal cortex. These same regions were recruited in 13 children aged 6-11 years when they listened to sections of a story describing a characters thoughts compared to sections of the same story that described the physical context. A distinct region in the posterior superior temporal sulcus was implicated in the perception of biological motion. Change in response selectivity with age was observed in just one region. The right temporo-parietal junction was recruited equally for mental and physical facts about people in younger children, but only for mental facts in older children.

Neural evidence for "intuitive prosecution": The use of mental state information for negative moral verdicts

Moral judgment depends critically on theory of mind (ToM), reasoning about mental states such as beliefs and intentions. People assign blame for failed attempts to harm and offer forgiveness in the case of accidents. Here we use fMRI to investigate the role of ToM in moral judgment of harmful vs. helpful actions. Is ToM deployed differently for judgments of blame vs. praise? Participants evaluated agents who produced a harmful, helpful, or neutral outcome, based on a harmful, helpful, or neutral intention; participants made blame and praise judgments. In the right temporo-parietal junction (right TPJ), and, to a lesser extent, the left TPJ and medial prefrontal cortex, the neural response reflected an interaction between belief and outcome factors, for both blame and praise judgments: The response in these regions was highest when participants delivered a negative moral judgment, i.e., assigned blame or withheld praise, based solely on the agents intent (attempted harm, accidental help). These results show enhanced attention to mental states for negative moral verdicts based exclusively on mental state information.

Combining motion planning and optimization for flexible robot manipulation

Robots that operate in natural human environments must be capable of handling uncertain dynamics and underspecified goals. Current solutions for robot motion planning are split between graph-search methods, such as RRT and PRM which offer solutions to high-dimensional problems, and Reinforcement Learning methods, which relieve the need to specify explicit goals and action dynamics. This paper addresses the gap between these methods by presenting a task-space probabilistic planner which solves general manipulation tasks posed as optimization criteria. Our approach is validated in simulation and on a 7-DOF robot arm that executes several tabletop manipulation tasks. First, this paper formalizes the problem of planning in underspecified domains. It then describes the algorithms necessary for applying this approach to planar manipulation tasks. Finally it validates the algorithms on a series of sample tasks that have distinct objectives, multiple objects with different shapes/dynamics, and even obstacles that interfere with object motion.

Hierarchical Decision Theoretic Planning for Navigation Among Movable Obstacles

In this paper we present the first decision theoretic planner for the problem of Navigation Among Movable Obstacles (NAMO). While efficient planners for NAMO exist, they are challenging to implement in practice due to the inherent uncertainty in both perception and control of real robots. Generalizing existing NAMO planners to nondeterministic domains is particularly difficult due to the sensitivity of MDP methods to task dimensionality. Our work addresses this challenge by combining ideas from Hierarchical Reinforcement Learning with Monte Carlo Tree Search, and results in an algorithm that can be used for fast online planning in uncertain environments. We evaluate our algorithm in simulation, and provide a theoretical argument for our results which suggest linear time complexity in the number of obstacles for typical environments.

Cart pushing with a mobile manipulation system: Towards navigation with moveable objects

Robust navigation in cluttered environments has been well addressed for mobile robotic platforms, but the problem of navigating with a moveable object like a cart has not been widely examined. In this work, we present a planning and control approach to navigation of a humanoid robot while pushing a cart. We show how immediate information about the environment can be integrated into this approach to achieve safer navigation in the presence of dynamic obstacles. We demonstrate the robustness of our approach through long-running experiments with the PR2 mobile manipulation robot in a typical indoor office environment, where the robot faced narrow and high-traffic passageways with very limited clearance.

Planning with movable obstacles in continuous environments with uncertain dynamics

In this paper we present a decision theoretic planner for the problem of Navigation Among Movable Obstacles (NAMO) operating under conditions faced by real robotic systems. While planners for the NAMO domain exist, they typically assume a deterministic environment or rely on discretization of the configuration and action spaces, preventing their use in practice. In contrast, we propose a planner that operates in real-world conditions such as uncertainty about the parameters of workspace objects and continuous configuration and action (control) spaces. To achieve robust NAMO planning despite these conditions, we introduce a novel integration of Monte Carlo simulation with an abstract MDP construction. We present theoretical and empirical arguments for time complexity linear in the number of obstacles as well as a detailed implementation and examples from a dynamic simulation environment.

Learning non-holonomic object models for mobile manipulation

For a mobile manipulator to interact with large everyday objects, such as office tables, it is often important to have dynamic models of these objects. However, as it is infeasible to provide the robot with models for every possible object it may encounter, it is desirable that the robot can identify common object models autonomously. Existing methods for addressing this challenge are limited by being either purely kinematic, or inefficient due to a lack of physical structure. In this paper, we present a physics-based method for estimating the dynamics of common non-holonomic objects using a mobile manipulator, and demonstrate its efficiency compared to existing approaches.

Navigation Among Movable Obstacles with learned dynamic constraints

In this paper we present the first planner for the problem of Navigation Among Movable Obstacles (NAMO) on a real robot that can handle environments with under-specified object dynamics. This result makes use of recent progress from two threads of the Reinforcement Learning literature. The first is a hierarchical Markov-Decision Process formulation of the NAMO problem designed to handle dynamics uncertainty. The second is a physics-based Reinforcement Learning framework which offers a way to ground this uncertainty in a compact model space that can be efficiently updated from data received by the robot online. Our results demonstrate the ability of a robot to adapt to unexpected object behavior in a real office scenario.