MIT Media Lab Projects

ML_TOPICS

Time perception is a fundamental component in our ability to build mental models of our world. Without accurate and precise time perception, we might have trouble understanding speech, fumble social interactions, have poor motor control, hallucinate, or remember events incorrectly. Slight distortions in time perception are commonplace and may lead to slight dyslexia, memory shifts, poor eye-hand coordination, and other relatively benign symptoms, but could a diminishing sense of time signal the onset of a serious brain disorder? Could time perception training help prevent or reverse brain disorders? This project is a series of experimental tools built to assist and increase human time perception. By approaching time-perception training from various perspectives, we hope to find a tool or collection of tools to increase time perception, and in turn discover what an increase in time perception might afford us.

Tools for Super-Human Time Perception

We have designed a novel system to promote kindness and gratitude. We leverage pervasive technologies to naturally embed gratitude inspiration in everyday life. Mobile sensor data is utilized to infer optimal moments for stimulating contextually relevant thankfulness and appreciation. We analyze the interplay between mood, contextual cues, and gratitude expressions.

"Kind and Grateful": Promoting kindness and gratitude with pervasive technology

In September 2014, 150 parents, engineers, designers, and healthcare practitioners gathered at the MIT Media Lab for the "Make the Breast Pump Not Suck!" Hackathon. As one of the midwives at our first hackathon said, "Maternal health lags behind other sectors for innovation." This project brought together people from diverse fields, sectors, and backgrounds to take a crack at making life better for moms, babies, and new families.

Make the Breast Pump Not Suck Hackathon

(Im)possible Baby is a speculative design project that aims to stimulate discussions about the social, cultural, and ethical implications of emerging biotechnologies that could enable same-sex couples to have their own, genetically related children. Delivering a baby from same-sex parents is not a sci-fi dream anymore, due to recent developments in genetics and stem cell research. In this project, the DNA data of a lesbian couple was analyzed using 23andme to simulate and visualize their potential children, and then we created a set of fictional, "what if" future family photos using this information to produce a hardcover album which was presented to the couple as a gift. To achieve more public outreach, we worked with the Japanese national television service, NHK, to create a 30-minute documentary film following the whole process, which aired in October 2015.

(Im)possible Baby

(void*) is a virtual lounge that unites the physical and digital, allowing people and a cast of virtual characters to interact with each other. In this novel gathering place, as in a nightclub, the interaction focuses mostly on group dynamics and body language, and little on verbal communication. Building on SWAMPED!, this installation continues our work in �Intentional Characters,� �Sympathetic Interfaces� and �Autonomous Cinematography.� It also adds �Dynamic Music Composition� as an integral part of the system. (void *) will premiere this summer at Siggraph �99. Today we will be showing a video and demonstrating the technology.

(void *): A Cast of Characters

Typically, 1-teacher schools are viewed as relatively backwards due to lack of resources, but research shows that teachers at such schools can have stronger, fuller, and deeper relationships with their students, and that the students can be more autonomous and in control of their own learning while still more supportive of and cooperative with their peers. Technology can provide connectivity, eliminating some resource constraints. We go further, using new learning methodologies and technologies to build upon the strengths of 1-teacher schools, creating models of learning environments to serve as exemplars not only for rural education, but also for large urban environments.

1-Teacher Schools

Varied important problems can be solved using surprisingly approximate arithmetic.  We've designed a co-processor for such arithmetic that provides 100,000 cores on a single standard chip, or 1,000 cores in a sub-watt mobile device.  We are exploring applications of such machines in image and video processing.  Cost can be under a penny per core, and compared to CPUs, improvements in speed and energy use can exceed 10,000x.

10,000x More Efficient Computing

The 1998 Everest Expedition was the first major scientific expedition to  Mount Everest.  We joined forces with the Boston Museum of Science, NASA,  Yale, the Explorer's Club, and a world-class climbing team, to investigate  questions of geology, climatology, and human physiology.
.

1998 Everest Expedition

Suppose that one could put a connected, portable computer into the hands of all the children of the world who lack access to the kind of education that would prepare them for participation in a modern, knowledge-based society. Would you use the computers to put the children through the standard curriculum of "developed" countries? This project—a complex undertaking—sets out to design radically different approaches. Some of its components, such as theoretical studies of the epistemology of learning, extend work we have pursued in the past. A component not undertaken before is developing a scientific basis for the design and implementation of a whole essentially new curriculum—or more likely, the process through which such a curriculum will evolve. Experiences such as the pilot program, which implemented a 1:1 computer infrastructure in a small rural community in Costa Rica, are been developed and studied.

1:1 Laptops: Learning with the Hundred-Dollar Laptop

Touch screen systems allow users to interact directly with content displayed on a 2-D surface. However, current systems have severe limitations in usability because they do not preserve the familiar mechanical response of traditional writing and drawing tools. Furthermore, current touch-screen technologies can often require extensive modifications to the drawing/writing substrate, which make them expensive and impractical for large drawing surfaces or entire desktops. We are developing a touch-screen system based on an effect called Frustrated Total Internal Reflection (FTIR). This technique involves coupling light into a pane of glass and then detecting contact to the glass by measuring the absorption of the light by the object in contact, using a linear array of photodetectors adhered to the glass. An LCD screen or other video display can then be fitted behind the glass to produce an appropriate response. All hardware is inside the glass�there is no need to mount anything on the side where the interaction takes place. We believe this system will be a practical and economical way to detect contact with brushes, felt pens, and fingers across large interactive surfaces.

2-D Input Device Based on Frustrated Total Internal Reflection

20 Day Stranger is a mobile app that creates an intimate and anonymous connection between you and another person. For 20 days, you get continuous updates about where they are, what they are doing, and eventually even how they are feeling, and them likewise about you. But you will never know who this person is. Does this change the way you think about other people you see throughout your day, any one of which could be your stranger?

20 Day Stranger

DuoSkin is a fabrication process that enables anyone to create customized functional devices that can be attached directly to the skin. Using gold metal leaf, a material that is cheap, skin-friendly, and robust for everyday wear, we demonstrate three types of on-skin interfaces: sensing touch input, displaying output, and wireless communication. DuoSkin draws from the aesthetics found in metallic jewelry-like temporary tattoos to create on-skin devices which resemble jewelry. DuoSkin devices enable users to control their mobile devices, display information, and store information on their skin while serving as a statement of personal style. We believe that in the future, on-skin electronics will no longer be black-boxed and mystified; instead, they will converge towards the user friendliness, extensibility, and aesthetics of body decorations, forming a DuoSkin integrated to the extent that it has seemingly disappeared.Credits:Cindy Hsin-Liu Kao, Asta Roseway*, Christian Holz*, Paul Johns*, Andres Calvo, Chris SchmandtMIT Media Lab in collaboration with Microsoft Research*

DuoSkin

Multi-layer 3-D microfluidic circuits are the driving platform for micro-total analysis systems in the current era. The traditional fabrication for these
fluidic circuits is planar. For building multi-layer structures a precise
alignment for multiple layers is required, which makes the fabrication very difficult. We have invented a flow-based extrusion process for multi-layer microfluidic circuits in polymer substrates. The channels can be as small as 100nm with multi-materials embedded in the substrate. Such a direct-write 3-D process will enhance testing and prototyping of microfluidic circuits and enable fabrication of true 3-D structures.

3-D Rapid Microfluidic Circuit Fabrication

The "300M IT Edition" project is a collaboration between the Media Lab, MIT's Aero/Astro Department, DaimlerChrysler, and Motorola. Using the project platform�a fully instrumented version of the Chrysler 300M�we collected data on the driver's mental workload and emotional state. Several demonstrations showed examples of how an intelligent cockpit could improve driving. For example, it could remind drivers to drive their best, and make bicycling near the vehicle safer.

300M IT Edition

The crystal oscillator inside a quartz wristwatch vibrates at 32,768 times per second. This is too fast for a human to perceive, and it's even more difficult to imagine its interaction with the mechanical circulation of a clock. 32,768 Times Per Second is a diagrammatic, procedural, and fully functional sculpture of the electro-mechanical landscape inside a common wristwatch. Through a series of electronic transformations, the signal from a crystal is broken down over and over, and then built back up to the human sense of time.

32,768 Times Per Second

This project presents a new approach for the design and fabrication of the acoustic instrument, using digital fabrication technologies, and specifically 3D printing. A concert flute was 3D printed, including (almost) all of its mechanisms, without the need to assemble the moving parts. In order to fulfill this challenge a new design of the instrument was made. Instead of proposing the machine can easily replace the human craft - we use this research to examine the possible contributions of rapid prototyping to the design and fabrication of traditional instruments.

3D Printed Flute

Functionally graded materials—materials with spatially varying composition or microstructure—are omnipresent in nature. From palm trees with radial density gradients, to the spongy trabeculae structure of bone, to the hardness gradient found in many types of beaks, graded materials offer material and structural efficiency. But in man-made structures such as concrete pillars, materials are typically volumetrically homogenous. While using homogenous materials allows for ease of production, improvements in strength, weight, and material usage can be obtained by designing with functionally graded materials. To achieve graded material objects, we are working to construct a 3D printer capable of dynamic mixing of composition material. Starting with concrete and UV-curable polymers, we aim to create structures, such as a bone-inspired beam, which have functionally graded materials. This research was sponsored by the NSF EAGER award: Bio-Beams: FGM Digital Design & Fabrication.

3D printing of functionally graded materials

An autostereoscopic (no glasses) 3D display engine is combined with a "Pepper's Ghost" setup to create an office chair that appears to contain a remote meeting participant. The system geometry is also suitable for other applications, such as tabletop or automotive heads-up displays.

3D Telepresence Chair

The project explores how to best engage design collaborators in an ongoing dialogue following a design charette process. A video record of the charette is recorded, parsed, posted, and transformed into an active web of meaning so that salient ideas and themes can be followed through the many phases of discussion. Initial processing is human intensive; however the intention is to apply common-sense technology in the future.

3D: Digital Dialogs for Design

The 3DprintedClock project is the result of ready-assembled 3D printed computational mechanisms, and is related to research in the fields of rapid prototyping and digital fabrication. The clock was modeled in CAD software after an existing clock, and uses a weight and a pendulum to keep track of time. The CAD model was created according to the specifications of the 3D printer, assuring sufficient gaps and clearances for the different parts. In addition, support material, drainage, and perforations were added to allow for excess support material being removed after printing. The 3DprintedClock is intended to demonstrate the superior capabilities of 3D printing as a fabrication process. It should contribute toward the future use of 3D printers to replace injection molding and expensive tooling processes, and allow for on demand, customized, and �greener� consumer products.

3DprintedClock

Many of our online interactions take place in community spaces. We keep track of friends and share pictures on Facebook, chatter with friends on Twitter, and participate in discussions on online forums. But how do the design choices we make impact the kinds of social spaces that develop? To better understand this relationship, we conducted a study of a discussion forum with a particularly unusual design: 4chan.org. Perhaps best known for its role in driving Internet culture and its involvement with the "Anonymous" group, we believe 4chan's design plays a large role in its success, despite its counter-intuitiveness. In our first paper exploring this area, we quantify 4chan's ephemerality (there are no archives; most posts are deleted in a matter of minutes) and anonymity (there are no traditional user accounts, and most posts are fully anonymous) and discuss how the community adapts to these unusual design strategies.

4chan and /b/: Anonymity and Ephemerality

4K/8K Comics applies the affordances of ultra-high-resolution screens to traditional print media such as comic books, graphic novels, and other sequential art forms. The comic panel becomes the entry point to the corresponding moment in the film adaptation, while scenes from the film indicate the source frames of the graphic novel. The relationships among comics, films, social media, parodies, and other support materials can be navigated using native touch screens, gestures, or novel wireless control devices. Big data techniques are used to sift, store, and explore vast catalogs of long-running titles, enabling sharing and remixing among friends, fans, and collectors.

4K/8K Comics

Is it possible to create passive displays that respond to changes in viewpoint and incident light conditions? Holograms and 4D displays respond to changes in viewpoint. 6D displays respond to changes in viewpoint as well as surrounding light. We encode the 6D reflectance field into an ordinary 2D film. These displays are completely passive and do not require any power. Applications include novel instruction manuals and mood lights.

6D Display

The 8D Display combines a glasses-free 3D display (4D light field output) with a relightable display (4D light field input). The ultimate effect of this extension to our earlier BiDi Screen project will be a display capable of showing physically realistic objects that respond to scene lighting as we would expect. Imagine a shiny virtual teapot in which you see your own reflection, a 3D model that can be lighted with a real flashlight to expose small surface features, or a virtual flashlight that illuminates real objects in front of the display. As the 8D Display captures light field input, gestural interaction as seen in the BiDi Screen project is also possible.

8D Display

8K Time into Space is a user interface for a video exploration system with an 8K display. 8K is an ultra high-definition video system and it can present a huge amount of visual content on one display. In our system, video thumbnails with shifted playback time in chronological order are spaced out like tiles. The time range of a scene that a viewer wants to check can be adjusted with a touch interface, and resolution of the thumbnails is changed depending of the range. 8K Time into Space aims to provide responsive and intuitive experiences for video consumption.

8K Time into Space

Archived TV programs evoke earlier times. This application combines a video and music archive with an immersive screen and a simple user interface suitable for everyone, from children to the elderly, to create a "Time Machine" effect. The only key for exploring is the user's age. People can enjoy over 1,300 TV programs from the last seven decades without having to do tedious text searches. This  catalogue intuitively guides the user with an image array (64 different videos on one screen at the same time) that simplifies navigation and makes it immediate, rather than referencing it to previous screens.

8K Time Machine

People cannot type as fast as they think, especially when faced with the constraints of mobile devices. The focus of this project is developing an alternative approach to predictive text entry using Open Mind Common Sense.

A Common-Sense Approach to Predictive Text Entry

We propose cutting as a novel paradigm for ad hoc customization of printed electronic components. As a first example we created a printed, capacitive, multi-touch sensor, which can be cut by the end-user to modify its size and shape. This very direct manipulation allows the end-user to easily make real-world objects and surfaces touch-interactive, to augment physical prototypes and to enhance paper craft. We contribute a set of technical principles for the design of printable circuitry that makes the sensor more robust against cuts, damages, and removed areas. This includes novel physical topologies and printed forward error correction. A technical evaluation compares different topologies and shows that the sensor remains functional when cut to a different shape.

A Cuttable Multi-Touch Sensor

Users have high-level goals when they browse the Web or perform searches.  However, the two primary user interfaces positioned between users and the Web�Web browsers and search engines�have very little interest in users� goals.  Present-day Web browsers provide only a thin interface between users and the Web, and present-day search engines rely solely on keyword matching. This project leverages large knowledge bases of semantic information to provide users with a goal-oriented Web browsing experience. By understanding the meaning of Web pages and search queries, this project demonstrates how Web browsers and search engines can proactively suggest content and services to users that are both contextually relevant and personalized. This project presents (1) Creo, a Programming by Example system that allows users to teach their computers how to automate interactions with their favorite Web sites by providing a single demonstration; (2) Miro, a data detector that matches the content of a Web page to high-level user goals, and allows users to perform semantic searches; and (3) Adeo, an application that streamlines browsing the Web on mobile devices, allowing users to complete actions with a minimal amount of input and output.

A Goal-Oriented User Interface for Personalized Semantic Search

The Kit-of-No-Parts is an approach to crafting electronics rather than designing discrete components. The collection of recipes and ingredients on the Kit-of-No-Parts website describes how to build electronics from a wide variety of conductive and non-conductive materials using a range of traditional and contemporary craft techniques.

A Kit-of-No-Parts

The SEC is an extension to the free open-source program EyesWeb�that contains a large number of machine learning and signal processing algorithms that have been specifically designed for real-time pattern and gesture recognition.  All the algorithms within the SEC are encapsulated as individual blocks, allowing the user to connect the output of one block to the input of another to create a signal flow chain.  This allows a user to quickly build and train their own custom gesture recognition system, without having to write a single line of code or explicitly understand how any of the machine learning algorithms within their recognition system work.

A Machine Learning Toolbox for Musician Computer Interaction

Mood classification is very subjective. There have been many psychological studies performed throughout the past century relating to music and emotion, and as a result there exist many different representations of human emotion. This work proposes to utilize the best possible psychological model of emotion by incorporating the findings of these studies into an innovative front-end for a digital music library where music can be queried, browsed, and explored by mood, rather than artist, album, or genre. With the use of state-of-the-art audio and textual analysis tools this work proposes to automatically and accurately classify music by mood.

A Mood-Based Music Classification and Exploration System

One of the fundamental goals of Montessori education is to create productive, stress-free educational environments for children. In addition to traditional methods of observation, we argue that teachers would benefit from tools that could provide supplemental data identifying stress responses in students using psychophysiological data. The child-suited wearable device we have designed incorporates sensors that track signs linked to emotional and sympathetic responses, such as heart rate variability and electro-dermal activity. Through these data points, teachers and parents can better understand the child's emotional responses to activities and social interactions at school, and tailor programs to support wellbeing and stress reduction.

A Multi-Sensor Wearable Device for Analyzing Stress Response in Preschool Classrooms

We have developed tightly coupled sensor network that tries to mimic some of the sensory characteristics of a natural skin. Developed in a flexible substrate (Kapton Polyimide), this array forms a flexible surface of small, networked sensor elements. Each node is capable of sensing four channels of pressure, two axes of strain, sound, temperature, ambient light, and two channels of whisker mechanoreception. By seamlessly blending sensing, computation, and local and global networking, our "skin" is ultimately self-configuring and scalable�it can be built as large as desired and signals can be extracted from the entire array at any node.

A Scalable Networked Smart Sensate Skin

We propose a flexible light-field camera architecture that represents a convergence of optics, sensor electronics, and applied mathematics. Through the co-design of a sensor that comprises tailored, angle-sensitive pixels and advanced reconstruction algorithms, we show that, contrary to light-field cameras today, our system can use the same measurements captured in a single sensor image to recover either a high-resolution 2D image, a low-resolution 4D light field using fast, linear processing, or a high-resolution light field using sparsity-constrained optimization.

A Switchable Light-Field Camera

A design project done in collaboration with the MIT Media Lab and the Laboratorio para la Ciudad (Laboratory for the City), Mexico City's experimental office for civic innovation and urban creativity, A-pops is a networked learning experience across Mexico City that supports young learners in engaging in emergent and playful opportunities in and beyond their local communities. In line with the "Playful City" goal, this project aims to embed playful learning experiences across Mexico City that are creative, collaborative, and public, by leveraging existing public spaces throughout neighborhoods and micro-communities across the city. By embedding a variety of playful learning experiences across a variety of locations, a wide range of learners have the ability to easily and socially engage in transformative experiences that support key skills in design, collaboration, creativity, programming, and learner agency.

A-pops

AAGO is a mobile app for Apple devices focusing on media creation, organization and sharing. Targeted to teens, the app aims to help them document group or individual creative projects (filmmaking, tech or art projects, citizen journalism) by creating "stories" made up of mobile photos, videos, and audio clips, which can then be arranged and exported to the web.

AAGO: Mobile Media Diaries for Youth Citizens Journalists

About Us examines technology workplaces from a feminist perspective. The objective is to mobilize workers to collect simple gender-specific data, such as the male-female ratio of their workplace. This data will be made available to others to inform performances, wearables, and games. The idea is to create collaboratively a data infrastructure to support projects and perspectives that shed light on the complicated issue of gender imbalance in spaces of high-tech labor.

About Us

AboutFace is a user-dependent system that is able to learn patterns and  discriminate the different facial movements characterizing confusion and  interest. The system uses a piezoelectric sensor to detect eyebrow  movements and begins with a training session to calibrate the unique values  for each user. After the training session, the system uses these levels to  develop an expression profile for the individual user.  The system has many  potential uses, ranging from computer and video-mediated conversations to  interactions with computer agents.  This system is an alternative to using  camera-based computer vision analysis to detect faces and expressions.  Additionally, when communicating with other people, users of this system  also have the option of conveying their expressions anonymously by wearing  a pair of glasses that conceals their expressions and the sensing device.

AboutFace

We are studying how social accountability information about companies affects consumer behavior. We are also developing Web-based tools to help consumers learn about and act on such information.

Account-Ability

Technical devices are generally considered to be compliant instruments that are fully controllable tools, and from this perspective, humans have total power to operate machines at will. However, technologies often act back on humans, reversing the intended command direction. The Acoustic Chase project examines processes in which machines are intentionally provided with �degrees of freedom� not fully controllable by people. This will partially invert the relationship of human and tool, and create a system of balanced power. Expanding on the Haptic Opposition project, which combined touch and vision to create a human-machine interaction wherein the apparatuses could push back against human motion, Acoustic Chase will extend the concept of machines with a �free� acting will, but use sound as the primary output. The sensation of sound is direct; it offers quick information flow and a great variety of expressions. It is also difficult to escape: people can�t "hear away" or "close their ears." Tracking human body movements will provide an input channel. In combination with spatial sound, Acoustic Chase will enable intuitive human reactions, like discovering, playing, or fleeing.

Acoustic Chase

This work seeks to apply acoustical analysis to a hierarchical model of discourse structure so as to allow for the division of an audio recording into layers of detail. This "outline" structure is then used in interactive presentations, which allow a listener to browse at various levels of detail.

Acoustical Cues to Discourse Structure

We are interested in building systems which can learn to connect natural language to objects and actions in video. In our current experiment, we have created videos of a person moving objects around on a table top. Naive subjects have been asked to describe verbally the video sequences. Our goal is to create a learning system which is able to "ground" the meaning of words and phrases in terms of observations found in the video. Applications of this work include verbal control of robots and natural-language-based access to video archives.

Acquiring Verbs from Speech and Video

Agents (human or computer) provide their users with recommendations. This project investigates whether the type of agent making the recommendations (human vs. computer, collaborative vs. individual) influences the decision strategy that people utilize. Specifically, we examine whether the type of agent used influences decision makers' price/quality tradeoffs, risk attitude, willingness to compromise or to follow their gut feeling, or their likelihood of becoming more sensitive to social influences.

Acting on Advice from Computer Agents

Action Path is a mobile app to help people learn about and engage with issues in their community. The app uses push notifications tied to geography that invite people to provide meaningful feedback on nearby issues as they traverse the city. Most platforms for civic engagement, whether online or offline, are inconvenient and disconnected from the source of the issues they are meant to address. Action Path addresses barriers to effective civic engagement by inviting people's input, converting individual actions into collective action, and providing context and a sense of efficacy.

Action Path

Activ8 is a system of three short games: See-Saw, a balancing game for Google Glass; Jump Beat, a music beat matching game for Google Glass; and Learning to Fly, a Kinect game where two users keep a virtual bird in the air by flapping their arms in sync. Recent epidemiological evidence points at sitting as being the most common contributor to an inactive lifestyle.  We aim to offer a starting point towards designing and building an understanding about how "physical casual games" can contribute to helping address the perils of sitting.

Activ8

Sleepiness is known to decrease driving performance, resulting in one of the major causes of fatal car accidents. An effective countermeasure to drowsiness is to increase activation, as activity and moderate exercise increase alertness. We are currently using redesigned accelerator pedals to integrate some form of physical activity while driving to maintain alertness levels in drivers.

Active Driver

Active Essays were forms of narrative expression in which computational objects were integrated with text, graphics, and video. For example, a simulation of a scientific concept could be integrated with text descriptions of the concept, so that "readers" actively explore the concept.

Active Essays

Active Messenger filters email messages according to dynamic rules, and automatically routes messages to a variety of wired and wireless delivery channels. It observes traffic from a user by knowing which devices have been used to originate or to respond to messages, recent log-ins, and caller ID when checking voice or email over the phone. Its goal is to make sure that desired messages always reach the subscriber, while paring messages down when the user is less reachable. Active Messenger also acts as a proxy, hiding the identity of the multiple device addresses at which the subscriber may be found. More than just a router, Active Messenger is a dynamic process which monitors a message's progress through various channels over time.

Active Messenger

We have developed an extremely low-power and low-cost wireless sensor network aimed at applications in asset tracking and ubiquitous activity monitoring. In addition to identifying an object, these nodes (termed active radio-frequency identification [RFID] tags) employ sensors to monitor its state, enabling new applications in fields like security, home automation, and supply-chain management. Although they contain a battery, these tags are not limited by it: by minimizing power consumption and quasi-passively waking on diverse stimuli (changes in light, RF carrier presence, shock or acceleration, and sound), they can last for years. Furthermore, their low cost and small size make them a good candidate for large-scale experiments at the intersection of RFID and wireless sensor networks.

Active RFID Tags for Security and Supply-Chain Management

Smart agents represent users by creating models of particular users, and the agents' recommendations are based on these models. But models are rarely complete; this problem is particularly acute when there are changes either in the tastes and preferences of the individual, or in the marketplace (e.g., new products). Our agent attempts to solve this by actively trying to learn a user's preferences. The active agent balances two goals: to be immediately useful and to make high-quality recommendations ("selling"); and to learn more about the user. These recommendations have a lower quality, but also a higher "up-side" potential ("learning").

Active-Learning Smart Agents

The activeShelves are a networked system of surfaces, servers, and displays that give real-time data about a collection of physical objects. Shelf surfaces are embedded with radiofrequency (RF) readers to track tagged containers. The system is connected to a server, and can be used as a relational database for physical objects. The user gets information directly about usage, location, and status of objects on activeShelves, or can query remotely via the web. The activeShelves can display simple text and graphics on LED displays. Based on our work with TouchCounters, activeShelves continues the exploration of enhancing typical organizational systems. The new system senses using RFID tags instead of mechanical magnetic contacts. The system's data-entry components and o utput display LEDs are located on the shelves instead of just the physical containers. The activeShelves system acts as a relational database instead of just tracking usages of individual containers.

activeShelves: Interactive Storage Shelf System

The Actuated Workbench is a device that uses magnetic forces to move objects on a table in two dimensions. It is intended for use with existing tabletop tangible interfaces, such as Sensetable, providing an additional feedback loop for computer output and helping to resolve inconsistencies that otherwise arise from the computer�s inability to move objects on the table. Current developments include the physical synchronizing of two distributed tables for remote collaboration, and the use of actuation to program physical constraints into a simulation, allowing users to quickly set up a physical system that determines possible parameters for the underlying computational simulation.

Actuated Workbench

String instruments are typically close-miked, to capture qualitative presence in recordings. This has been impossible for woodwind instruments, since each note comes from a different place on the instrument. We are developing microphones that can overcome this difficulty for woodwind instruments. By embedding the microphone with proximity detection, pitch detection, and tonal analysis, a microphone can continually adapt the audio signal. The instrument�s natural voicing will be preserved when close-miked, making it possible to record woodwind instruments with more detail.

Adaptive Microphones

In this project, we propose a new Bayesian receiver for signal detection in flat-fading channels. First, the detection problem is formulated as an inference problem in a hybrid dynamic system that has both continuous and discrete variables. Then, an expectation propagation algorithm is proposed to address the inference problem. As an extension of belief propagation, expectation propagation efficiently approximates a Bayesian estimation by iteratively propagating information between different nodes in the dynamic system and projecting exact messages into the exponential family. Compared to sequential Monte Carlo filters and smoothers, the new method has much lower computational complexity since it makes analytically deterministic approximations instead of Monte Carlo approximations. Our simulations demonstrate that the new receiver achieves accurate detection without the aid of any training symbols or decision feedbacks. Future work involves joint decoding and channel estimation, where convolutional codes are used to protect signals from noise corruption. Initial results are promising.

Adaptive, Wireless, Signal Detection and Decoding

The Additive Lathe is a 3D printer that prints onto rotating objects. It has two degrees of freedom: the first degree is the x-axis that moves with a rack and pinion mechanism. The second degree is the rotation of the object. Unlike regular lathes, in the Additive Lathe the object can spin in both directions and can spin in steps due to the stepper motor that spins the spindle. The additive lathe can produce objects that are difficult to make with regular 3D printer like threads and springs.

Additive Lathe

Our initial experiments in spark electrosintering fabrication have demonstrated a capacity to solidify granular materials (35-88 micron soda ash glass powder) rapidly using high voltages and power in excess of 1 kW. The testbed high-voltage setup comprises a 220V 60A variable autotransformer and a 14,400V line transformer. There are two methods to form members using electrosintering: the one-electrode drag (1ED) and two-electrode drag (2ED) techniques. The 1ED leaves the first electrode static while dragging the second through the granular mixture. This maintains a live current through the drag path and increases the thickness of the member due to the dissipation of heat. Large member elements have been produced with a tube diameter of around 0.75". The 2ED method pulls both electrodes through the granular mixture together, sintering the material between the electrodes in a more controlled manner.

Additive Manufacturing in Glass: Electrosintering and spark gap glass

Users don�t want to browse the Web on mobile devices, they want to complete actions. Adeo is an application that streamlines accessing the Web on mobile devices. Users can teach a computer how to interact with their favorite Web sites by demonstrating the actions in a Web browser, and then play these actions back with a single click on their mobile device. Adeo reduces complex procedures to the minimal amount of required input and output, improving both the efficiency and usability of mobile browsing.

Adeo: A Streamlined User Interface for Accessing the Web on Mobile Devices

This project explores the contribution of advanced audio systems to live performance, their design and construction, and their integration into the theatrical design process. We look specifically at innovative input and control systems for shaping the analysis and processing of live performance; and at large-scale output systems which provide a meaningful virtual abstraction to DSP in order to create flexible audio systems that can both adapt to many environments and achieve a consistent and precise sound field for large audiences.

Advanced Audio Systems for Live Performance

Computer modeling and simulation are changing the nature of scientific investigation, by enabling researchers to pose new kinds of questions and explore phenomena in ways that were not possible just a short time ago. Just as the technological revolution continues to influence the practice of scientific research, it also presents opportunities to change the way that science is taught in the classroom. The Adventures in Modeling Project introduced students and teachers to the process of designing, creating, and analyzing their own models of complex, dynamic systems using StarLogo. Our aims were to educate and motivate teachers to transform the way that they teach science and to engage students in authentic science practice by giving them the tools and the ability to pose, investigate, and answer their own questions. Through model creation, and participation in a variety of off-computer activities, teachers and students learned the science of modeling to develop a deeper understanding of patterns and processes in the world.

Adventures in Modeling

Suitable for anywhere a "Pepper's Ghost" display could be deployed, this display adds 3D with motion parallax, as well as optically relaying the image into free space such that gestural and haptic interfaces can be used to interact with it. The current version is able to display a person at approximately full-size.

Aerial Light-Field Display

Affect as Index is a tool that takes group physiological data as input, aggregates it across different demographic dimensions, and attaches them to media content. Users can review videotaped or prerecorded events by clicking on points of interest in a physiological graph. This software addresses two challenges: the difficulty of expressing and sharing emotions with others, and the laborious task of monitoring interpersonal interactions within natural settings. For the former, groups interested in discussing shared and dissimilar emotions evoked during experiences can use this tool to place context around their dialogue. For the latter, "meaningful moments" observed within natural interactions can be marked and superimposed on the physiological data collected. In this way, affect and observations of affect can be used to index group-level significant moments that occur within volumes of video data.

Affect as Index

The aim of this project is to build a database of natural speech showing a range of affective variability. It is an extension of our ongoing research focused on building models for automatic detection of affect in speech. At a very basic level, training such systems requires a large corpus of speech containing a range of emotional vocal variation. A traditional approach to this research has been to assemble databases where actors have provided the affective variation on demand. However, this method often results in unnatural sounding speech and/or exaggerated expressions. 

We have developed a prototype of an interactive system that guides a user through a question and answer session. Without any rehearsals or scripts, the user navigates through touch and spoken language an interface guided by embodied conversational agents which prompt the user to speak about an emotional experience. Some of the issues we are addressing include the design of the text and character behavior (including speech and gesture) so as to obtain a convincing and disclosing interaction with the user.

Affect in Speech:  Assembling a Database

The "Affective Carpet" is a soft, deformable surface made of cloth and  foam, which detects continuous pressure with excellent sensitivity and  resolution.  It is being used as an interface for projects in affective expression, including as a controller to measure a musical performer's direction and intensity in leaning and weight-shifting patterns.

Affective Carpet

Affective Circles combines the Chat Circles 2 interface with affective information, including skin conductivity data, to enhance online communication.

Affective Circles

The Affective Mirror is an attempt to build a fully automated system that intelligently responds to a person's affective state in real time. Current work is focused on building an agent that realistically mirrors a person's facial expression and posture. The agent detects affective cues through a facial-feature tracker and a posture-recognition system developed in the Affective Computing group; based on what affect a person is displaying, such as interest, boredom, frustration, or confusion, the system responds with matching facial affect and/or posture. This project is designed to be integrated into the Learning Companion Project, as part of an early phase of showing rapport-building behaviors between the computer agent and the human learner.

Affective Mirror

This study attempts to examine humans' affective responses to superimposed sinusoidal signals. These signals can be perceived either through sound, in the case of electronically synthesized musical notes, or through vibro-tactile stimulation, in the case of vibrations produced by vibrotactile actuators. This study is concerned with the perception of superimposed vibrations, whereby two or more sinusoisal signals are perceived simultaneously, producing a perceptual impression that is substantially different than of each signal alone, owing to the interactions between perceived sinusoidal vibrations that give rise to a unified percept of a sinusoidal chord. The theory of interval affect was derived from systematic analyses of Indian, Chinese, Greek, and Arabic music theory and tradition, and proposes a universal organization of affective response to intervals organized using a multidimensional system. We hypothesize that this interval affect system is multi-modal and will transfer to the vibrotactile domain.

Affective Response to Haptic Signals

ASQ investigates ways to teach social-emotion skills to children interactively with toys. One of the  first goals is to help autistic children recognize expressions of emotion in social situations. The system  uses four "dwarfs" expressing sad, happy, surprise, and angry, and each  communicates wirelessly to  the system and detects which plush doll was selected by the child.  The computer plays short  entertaining video clips displaying examples of the four emotions and cues the child to pick a dwarf  that closely matches that emotion.  Future work includes improving the ability of the system to  recognize direct displays of emotion by the child.

Affective Social Quest

Predicting media effectiveness from automatically measured affective responses collected "in-the-wild" over the Internet. This project shows that it is possible to predict powerful measures of media effectiveness automatically from natural and spontaneous facial responses captured via the web.

Affective Storytelling

People naturally express frustration through the use of their motor skills. The purpose of the Affective Tangibles project is to develop physical objects that can be grasped, squeezed, thrown, or otherwise manipulated via a natural display of affect. Constructed tangibles include a PressureMouse, affective pinwheels that are mapped to skin conductance, and a voodoo doll that can be shaken to express frustration. We have found that people often increase the intensity of muscle movements when experiencing frustrating interactions.

Affective Tangibles

The Affective Tigger is a plush toy designed to recognize and react to certain emotinal behaviors of its playmate.  For example the toy enters a state of "happy," moving its ears upward and emitting a happy vocalization when it recognizes that the child has postured the toy upright and is bouncing it along the floor.  Tigger has five such states, involving recognizing and responding with an emotional behavior.  The resulting behavior Tigger demonstrates allows it to serve as an affective mirror for the child's expression.  This work involved designing the toy, and evaluating sessions of play with it with dozens of kids.  The toy was shown to successfully communicate some aspects of emotion, and to prompt behaviors that are interesting to researchers trying to learn about the development of human emotional skills such as empathy.

Affective Tigger

Recent findings in affective neuroscience and psychology indicate that human affect and emotional experience play a significant and useful role in human learning and decision-making. Most machine-learning and decision-making models, however, are based on old, purely cognitive models, and are slow, brittle, and awkward to adapt. We aim to redress many of these classic problems by developing new models that integrate affect with cognition. Ultimately, such improvements will allow machines to make smarter and more human-like decisions for better human-machine interaction.

Affective-Cognitive Framework for Machine Learning and Decision Making

Companies would like more new products to be successful in the marketplace, but current evaluation methods such as focus groups do not accurately predict customer decisions. We are developing new technology-assisted methods to try to improve the customer-evaluation process and better predict customer decisions. The new methods involve multi-modal affective measures (such as facial expression and skin conductance) together with behavioral measures, anticipatory-motivational measures, and self-report cognitive measures. These measures are combined into a novel computational model, the form of which is motivated by findings in affective neuroscience and human behavior. The model is being trained and tested with customer product evaluations and marketplace outcomes from real product launches.

Affective-Cognitive Product Evaluation and Prediction of Customer Decisions

AffQuake is an attempt to incorporate signals that relate to a player's affect into ID Software's Quake II in a way that alters game play. Several modifications have been made that cause the player's avatar within Quake to alter its behaviors depending upon one of these signals. In StartleQuake, when a player becomes startled, his or her avatar also becomes startled and jumps back. Quake changes the size of the player's avatar in relation to the user's response as well, representing player excitement by average skin conductance level, and growing the avatar's size when this level is high.  A taller avatar means the player can see further; however, it also makes him or her an easier target.

AffQuake

The AFK cookset is designed for the hungry role-playing gamer who can connect her food items, e.g., Spicy Wolf Dumplings, to her online cooking habits. By scanning in the food items, the video game physically adjusts a hot plate to cook the item for the correct amount of time. The virtual character then jubilantly announces the status of the meal to both the gamer and the other individuals playing online: �O la la my roasted raptor is about to be done!� When the food is ready, the system automatically puts the character in AFK (�Away From Keyboard�) mode to provide the gamer with a moment to eat. When the player resumes playing, he or she might just discover his or her character�s behavior is affected by the food consumed in real life�sluggish from overeating or alternately exuberant and energetic.

AFK Cookset

The AmEC Infrastructure is an attempt to mitigate the well-known problem of software reuse in the domain of agent-mediated electronic commerce. Most of the systems developed by the AmEC initiative share a set of requirements that, once met, could serve as the foundation of future electronic commerce systems. The AmEC Infrastructure will define both a framework for the development of new systems and a set of pre-built subsystems that could be incorporated into new electronic commerce agent systems.

Agent Mediated Electronic Commerce (AmEC) Infrastructure

Current experiments in agent software rely mostly on domain-specific applications that either have been programmed from scratch, or explicitly modified with an agent in mind. Is it possible to make a tool kit or protocol that would allow an agent to communicate and to control applications that have been constructed more conventionally? Can the agent "take the place" of the user in the interface? Can the agent have access to the application's data and behavior? Will commercial "inter-application communication" mechanisms suffice? What is the division of labor between the agent and the application? This work will explore these questions.

Agent-Application Communication

What would it mean to browse collaboratively? In business meetings or family situations, several people may be attending a browsing session, even if only one "controls" the browser. The success of browsing depends on satisfying the interests of the passive participants as well. We are working on an agent that will provide assistance to group browsing, representing the interests of the entire group, and performing "reconnaissance" in anticipation of what pages are likely to be of common interest. It features automatic detection of the presence of users, acquisition of information about the interests of users, automated "channel surfing," and dynamic display of the user profiles and explanation of recommendations.

Agents for Collaborative Browsing

Effective image annotation and retrieval is bound up with image use. In this project, annotation, retrieval, and use are integrated, facilitating the finding and using images. A proactive user-interface agent seeks chances for image annotation and retrieval in the context of the user's everyday work, using an agent that sit in the user's text editor or other application and continuously monitors typing. Searches are automatically performed from an image library, and images relevant to the current text can be inserted in a single operation. Descriptions of images for storytelling can be seamlessly employed as raw material for annotation. Common-sense knowledge about situations in which pictures are taken, described, or used can help provide semi-automatic annotation and indirect inference for retrieval. Our approach does not completely automate the annotation/retrieval process, but it does reduce user-interface overhead, leading to better-annotated image libraries and fewer missed opportunities for image use.

Agents for Integrated Annotation and Retrieval of Images

AgI connects isolated, rural farmers to markets, to knowledge, and to each other. Farmers will have access to market price information, weather data, agricultural extension services, buyers and intermediaries, and relevant experts. Importantly, a synchronous communication module enables farmers to establish voice and data communication with peer farmers from other rural communities. The system supports a geography-based user interface that respects a range of written literacy levels. For instance, all market price information includes text representations and machine-generated, spoken audio.

AgI: Agricultural Information System

The first of this project's two goals is to enable those with internalized, psychologically distressing states to speak their own stories out loud to others without fear. The second goal involves understanding the �architecture� of current public spaces, both physical and online, and designing the structure and interaction of a physically located audiospace project that enables and encourages citizens to respond easily, and to be responsible towards each other in urban public places. This project revolves around the design and building of a public installation to let people call from any touch-tone phone into a public space. At the site of this public space will be a sculpture of sorts�a physical tele-presence for the person calling. The person calling will actually speak out of the physical sculpture. A full duplex audio link will be opened up between the caller and the tele-presence sculpture via existing telephone infrastructures. People in the public space can talk with the remote caller by talking to the sculpture. This project brings the privacy and anonymity of online communication into a physical public space to facilitate emotional processing by individuals. Public disclosure of private information may open dialogues around topics otherwise private or dangerous. Individuals may take steps towards dealing with personal situations while simultaneously offering public citizens voices articulating broadly pertinent issues. People present at the site of the tele-presence sculpture will be called upon to be responsive towards others while simultaneously feeling less alone in facing issues they may also be dealing with personally. Public spaces will potentially become more vital, and private issues less alienating.

AgoraPhone

Drivers spend a significant amount of time multi-tasking while they are behind the wheel. These dangerous behaviors, particularly texting while driving, can lead to distractions and ultimately to accidents. Many in-car interfaces designed to address this issue still neither take a proactive role to assist the driver nor leverage aspects of the driver's daily life to make the driving experience more seamless. In collaboration with Volkswagen/Audi and the SENSEable City Lab, we are developing AIDA (Affective Intelligent Driving Agent), a robotic driver-vehicle interface that acts as a sociable partner. AIDA elicits facial expressions and strong non-verbal cues for engaging social interaction with the driver. AIDA also leverages the driver's mobile device as its face, which promotes safety, offers proactive driver support, and fosters deeper personalization to the driver.

AIDA: Affective Intelligent Driving Agent

A major problem for natural language interfaces is their inability to handle text whose meaning depends in part on context.  If a user asks his car radio to play "a fast song," or his calendar to schedule "a short meeting," the interpreter would have to accommodate vagueness and ambiguity to figure out what he meant based on what he said.  For it to understand what songs or events the speaker intended, it must make decisions that depend on assumed common knowledge about the world and language.  Our research presents two approaches for reducing uncertainty in natural language interfaces, by modeling interpretation as a plan recognition problem.

AIGRE: Natural Language Interface that Accommodates Vague and Ambiguous Input

This project is aimed at building a machine learning pipeline that will discover and predict links between the visible structure of villages and cities (using satellite and aerial imaging) and their inhabiting social networks. The goal is to estimate digitally invisible villages in India and Sub-Saharan Africa. By estimating the social structure of these communities, our goal is to enable targeted intervention and optimized distribution of information, education technologies, goods, and medical aid. Currently, this pipeline is implemented using a GPU-powered Deep Learning system. It is able to detect buildings and roads and provide detailed information about the organization of the villages. The output will be used to construct probabilistic models of the underlying social network of the village. Moreover, it will provide information on the population, distribution of wealth, rate and direction of development (when longitudinal imaging data is available), and disaster profile of the village.

AINA: Aerial Imaging and Network Analysis

Air Mobs creates a local mobile community to allow users to freely share Internet access among diverse carrier 3G and 4G data accounts. We have built an app where anyone can advertise that they have bits and battery to spare and are willing to let other Air Mob members tether to them. They might do this if they are near their data cap and either need a little more data, or have some they are willing to let others use before it expires. A website tracks the evolution of the community and posts the biggest donators and users of the system. To date, this app works on Android devices. It is designed to be open and community-based.  We may experiment with market credits for sharing airtime and adding other devices and features.

Air Mobs

aireForm is a dress of many forms that fluidly morph from one to another, animated by air, reflecting the shifting of our own personas. Pneumatic pillows transform the shape of sections of the dress, revealing new forms and evoking classic feminine silhouettes, from sleek to supple to striking.

aireForm

A smartphone based spectrometer design that is standalone and supported on a wireless platform. The device is low-cost and the power consumption is minimal making it portable to perform a range of studies in the field. Essential components of the device like the light source, spectrometer, filters, microcontroller and wireless circuits have been assembled in a housing that fits into a pocket and the entire device weighs 48 g. The device has a dedicated app on the smartphone to communicate, receive, plot and analyze spectral data. Validations of the device were carried out by demonstrating non-destructive ripeness testing in fruits. Ultra-Violet fluorescence from Chlorophyll present in the skin was measured across various apple varieties during the ripening process and correlated with destructive firmness tests. This demonstration is a step towards possible consumer, bio-sensing and diagnostic applications that can be carried out in a rapid manner.

Smartphone spectrometer for food sensing

"All Cows Eat Grass" is a mnemonic used in music instruction for A C E G�the notes on the white spaces of the bass clef. The All Cows Eat Grass project is an online platform that provides cost-effective, real-time private music lessons. The system connects music instructors and students using a low-latency audio and video link, provides motivational support to practice between lessons, and lowers the barrier to learn music with self-instruction material and self-evaluation musical games.

All Cows Eat Grass

AlphaWolf presents a synthetic wolf pack comprised of autonomous and semi-autonomous wolves who interact with each other much as real wolves do: forming dynamic social relationships based on their past experience. Several people can interact with this installation at the same time. Each participant affects the emotional state of her given wolf by howling, growling, or whimpering into a microphone. In addition, participants may encourage their wolves to interact with specific other wolves. The dynamics of how the wolves interact is determined by the wolves' internal state, their social position in the pack, and their previous experiences with their pack-mates. By letting participants "get inside the mind and body" of a wolf, we hope to give them a compelling opportunity to explore the meaning of social behavior. The installation features a suite of supporting technology, including evocative real-time computer graphics, autonomous cinematography, and dynamic scoring and sound design.

AlphaWolf

Urban noise pollution has been a problem since the days of Buddha. Walkmans help, but issues of both social and accoustic isolation have become more urgent with the popularity of the iPod. Addressing these issues may require a look at how recorded music devices work at a fundamental level. Ambient Addition is a Walkman-like device, built on a DSP core, that synthesizes music by sampling the sound around the listener, creating harmony and rhythm from the chaos and noise of the environment. By simultaneously opening music to incorporate the environment, but also turning the environment into music, the sound stays fresh and the listener is encouraged to explore new territory.

Ambient Addition

This project explores the use of physical fixtures and objects to display temporal information within an architectural space in a subtle and aesthetically pleasing way. Ambient displays present and communicate information through their changes in form, movement, sound, color, and light. Ambient displays are persistently connected to digital information sources, continually displaying changing information within the physical environment.

Ambient Displays

We design and evaluate an ambient blood glucose level visualization and feedback system that uses an Ambient Orb for diabetes self-care and social support.  The social support is provided by a friend or family member of an individual with diabetes. This research study was carried out with adult patients at Joslin Diabetes Center.

Ambient Displays for Social Support and Diabetes Management

Furniture is the infrastructure for human activity. Every day we open cabinets and drawers, pull up to desks, recline in recliners, and fall into bed. How can technology augment these everyday rituals in elegant and useful ways? The Ambient Furniture project mixes apps with the IKEA catalog to make couches more relaxing, tables more conversational, desks more productive, lamps more enlightening, and beds more restful. With input from Vitra and Steelcase, we are prototyping a line of furniture to explore ideas about peripheral awareness (Google Latitude doorbell), incidental gestures (Amazon restocking trash can and the Pandora lounge chair), pre-attentive processing (energy clock), and eavesdropping interfaces (Facebook photo coffee table).

Ambient Furniture

Traditional music production and studio engineering depends on dynamic range compression audio signal processors that precisely and dynamically control the gain of an audio signal in the time domain. This project expands on the traditional dynamic range compression model by adding a spatial dimension. Ambisonic Compression allows audio engineers to dynamically control the spatial properties of a three-dimensional sound field, opening new possibilities for surround-sound design and spatial music performance.

Ambisonic surround-sound audio compression

Amino is a design-driven mini-lab that allows users to carry out a bacterial transformation and enables the subsequent care and feeding of the cells that are grown. Inspired by Tamagotchis, the genetic transformation of an organism's DNA is performed by the user through guided interactions, resulting in a synthetic organism that can be cared for like a pet. Amino is developed using low-cost ways of carrying out lab-like procedures in the home, and is packaged in a suitcase-sized continuous bioreactor for cells.

Amino: A Tamagotchi for Synthetic Biology

Amphibian allows users to easily imprint digital functions onto common everyday physical objects. Amphibian is a low cost, low infrastructure system that enables users to choose their own physical objects and imprint onto them almost any standard interface functions that take place on a GUI desktop. The goal of Amphibian is to create a system that the common user can implement and operate so that we may learn more about the digital-physical object relationships people will form.

Amphibian

SCUBA diving as a sport has enabled people to explore the magnificent ocean diversity of beautiful corals, striking fish, and mysterious wrecks. However, only a small number of people are able to experience these wonders, as diving is expensive, mentally and physically challenging, needs a large time investment, and requires access to large bodies of water. Most existing SCUBA diving simulations in VR are limited to visual and aural displays. We propose a virtual reality system, Amphibian, that provides an immersive SCUBA diving experience through a convenient terrestrial simulator. Users lie on their torso on a motion platform with their outstretched arms and legs placed in a suspended harness. Users receive visual and aural feedback through the Oculus Rift head-mounted display and a pair of headphones. Additionally, we simulate buoyancy, drag, and temperature changes through various sensors.

Amphibian: Terrestrial SCUBA Diving Simulator Using Virtual Reality

Shape-changing materials have been part of sci-fi literature for decades. But if tomorrow we invent them how are we going to use them? Amphorm is a kinetic sculpture that resembles a vase. Since Amphorm is a dual citizen between the digital and the physical worlds, its shape can be altered both in the  physical world through hand gestures and in the digital realms through an iPad app. Through this project, we explore how the physical world could be synchronized to the digital world and how tools from both worlds can jointly alter dual-citizens.

Amphorm

We are building a new kind of agent that acts as a user's assistant in browsing the World Wide Web. Many current Web tools perform searches for the user, but our approach is to consider the search for information as a cooperative venture between the human user and an intelligent software agent. Rather than search a pre-indexed portion of the Web according to user-stated keywords, the agent, Letizia, infers interest implicitly from observing user actions and tries to stay just a few steps ahead of the user, searching the user's immediately accessible links dynamically.

An Advisory Agent for Web Browsing

This project examines how the expression granted by new musical interfaces can be harnessed to create positive changes in health and wellbeing. We are conducting experiments to measure EEG dynamics and physical movements performed by participants who are using software designed to invite physical and musical expression of the basic emotions. The present demonstration of this system incorporates an expressive gesture sonification system using a Leap Motion device, paired with an ambient music engine controlled by EEG-based affective indices. Our intention is to better understand affective engagement, by creating both a new musical interface to invite it, and a method to measure and monitor it. We are exploring the use of this device and protocol in therapeutic settings in which mood recognition and regulation are a primary goal.

Latent Lab