Thought Operated Random Access Memory Prototype for Implantable Memory Devices for Less Than $200

Ido Bachelet has previously created DNA nanorobots for nanomedicine surgery. His new research is thought operated digital random access memory

Thought-Operated Digital Random-Access Memory by Lee Ben-Amil and Ido Bachelet.

The capacity and reliability of biological memory could be exceeded by a constantly growing flux of information to remember and operate by. Yet, our memory is fragile and could be easily impaired, and the prevalence of memory disorders is increasing in correlation with the population’s mean age. As expected, auxiliary memory devices (such as writing pads and computers) are abundant but are operated indirectly using significant effort compared with biological memory. We report a working prototype of a simplified, 4 KB random-access memory (RAM) that can be written to or read from using thought and could be embedded more seamlessly than other artificial memory aids. The system analyses EEG signals to extract attention levels, which trained subjects can use to write messages into an RFID sticker, or read from it on a display. We describe basic modes of using memory by a single subject, emulate common forms of social communication using this system, and highlight new forms of social usage and allocation of memories that are linked to specific persons. This preliminary prototype highlights the technical feasibility and the possibilities of implantable thought-operated memory devices and could be developed further to provide seamless aid to people suffering from memory disorders in the near future.

The prototype described here is extremely preliminary in the sense that it is motivated by seamless embedding of memory without being seamless in itself. However, this is a technical barrier that is being tackled, or has been tackled successfully in some cases. RFID circuits such as the one used here are completely implantable, and their interference with existing devices such as pacemakers has been studied. The portability of other components of the system is being improved towards complete implants, or at least wearable or in patch form. EEG measurements themselves could be made using sensor pads or implantable sensors, eliminating the need for a carried EEG headset. Display of the content that is retrieved from the memory could be done by means of contact lens, or, less directly, on glasses such as Google glass. Eventually, a system similar to the one described here could be entirely implantable. Moreover, the capacity of 4 KB implemented here could certainly be increased in future designs.

The specific method of writing and reading from the device could be improved. Attention is a parameter that can be readily extracted from raw EEG signals, and our observations show that trained subjects could switch between desired levels of attention sufficiently for the system to recognize the appropriate function to be carried out. However, most (∼75%) messages contained at least 1 incorrect bit. This suggests that either there is a better parameter to guide the system by or that the short training provided in this study was not sufficient. Further experiments are underway to investigate additional parameters within EEG data that could be used and to evaluate the potential precision of their utilization.

Implantable memory devices raise their own issues of privacy, possibilities for unauthorized reading, and inadvertent manipulation. Physical proximity, as required in the described prototype, is an important protective factor but limits the social applications of such devices. To enable the full scale of uses, implantable memory devices should be designed with specific layers of security addressing these special challenges, such as interference from adjacent devices and other implants and potential attacks made against the person through the implanted device.

Comparing this work with other artificial memory devices introduced earlier shows the potential of a noninvasive prototype that can be used to store and share data between 2 or more persons and to use one mind or more as a “cloud” similar to sharing thoughts and memories in social networks or the Internet today. The ability to communicate in a standard network like NFC described here may offer a connection to other devices and may correlate to other languages in future work. In contrast, converting this prototype to an invasive one as other introduced implants may give other abilities of extending human memory and brain capacity capabilities that were not found in today’s implants.


9 subjects (5 women, 4 men, ages 18–43, average age 31.7 ± 10.9 years, σ2 118.27) were recruited. They chose controlling attention levels since it is a parameter that is well researched and tested in EEG data and already used in other works. There are some hardware and applications that already use it in different ways like computer games or for research. The reason for the four ranges is to create different letters and mode in a language of two digits (0/1) and to differ between read, write, and no request at all as explained in Figure 1.

1. Each subject underwent a short (average ∼15 min) phase of training of the system until they were able to achieve specific attention of one of four levels. These levels were defined based on a scale of 0–100% attention, and each was used to code a specific function:
0–29% read from memory,
30–59% baseline for “no action” and to differ between reading and writing,
60–79% write “0” to memory, and
80–100% write “1” to memory.

Each subject was allowed to achieve her/his own speed in switching between attention levels, with an average of 3.5 ± 1.2 s spent at each level at the end of the training phase. Attention levels defining the ends of the scale were achieved by experiencing passive activity versus a difficult mathematical problem as previously described. In the testing phase, the subjects were requested to read or write 0/1 by achieving the desired level of attention described above.

EEG data were acquired using a Neurosky Mindwave mobile plus kit headset that provides raw-sampled wave values (128 Hz or 512 Hz, depending on hardware), signal quality metrics, eSense attention meter values (0 to 100), and EEG band power values for delta, theta, alpha, beta, and gamma.

The Neurosky Mindwave kit can be bought for about $99 on eBay.

EEG signals were obtained from neurosky mobile algorithm analysis. The Attention meter algorithm (eSens) indicates the intensity of mental “focus” or “attention.” The value ranges from 0 to 100. The attention level increases when a user focuses on a single thought or an external object and decreases when distracted. Users can observe their ability to concentrate using the algorithm. In educational settings, attention to lesson plans can be tracked to measure their effectiveness in engaging students. In gaming, attention has been used to create “push” control over virtual objects.

eSense Attention meter indicates the intensity of a user’s level of mental “focus” or “attention,” such as that which occurs during intense concentration and directed (but stable) mental activity. Its value ranges from 0 to 100. Distractions, wandering thoughts, lack of focus, or anxiety may lower the Attention meter level. For each different type of eSense (i.e., Attention and Meditation), the meter value is reported on a relative eSense scale of 1 to 100. On this scale, a value between 40 and 60 at any given moment in time is considered “neutral” and is similar in notion to “baselines” that are established in conventional brainwave measurement techniques (though the method for determining a ThinkGear baseline is proprietary and may differ from other methods).

The signals were broadcast via Bluetooth to a controller for processing and classification. We used an Arduino Uno device connected to BlueSMiRF silver Bluetooth antenna ($28), which translated the signals from the mindwave mobile headset device using a custom-written code. To process and classify the signals, an additional code was written using Arduino language (based on C/C++). The base program handles the Attention signals and determines the levels to classify. An NFC (near field Communication) Reading/Writing antenna shield (13.56 MHz band) was connected to the controller. A Mifare classic RFID tag with a 4 KB memory storage (about 77 cents to $1) was used to store the data written or to broadcast the data when reading. Arduino and NFC antenna shields were connected to a DELL I5–4200U (2.3 GHz/4 GB RAM) laptop with windows 7 operating system which was used as display monitor.