Questions related to Human Performance
Students should evaluate the aesthetics (e.g. "this room is beautiful") of a working space / room and what they think would be their performance (e.g. "I'm very productive in this room") in this room, based on a picture (which they see in a survey).
How to measure the subjective aesthetics and the subjective performance (of a Person) in it?
Which are the relevant and up to date publications and thoughts on the topic?
I have packaged the new Caltech test set annotations provided by Shanshan Zhang in https://www.mpi-inf.mpg.de/departments/computer-vision-and-machine-learning/research/people-detection-pose-estimation-and-tracking/how-far-are-we-from-solving-pedestrian-detection/ to .vbb file format to use the evaluation code provided in http://www.vision.caltech.edu/Image_Datasets/CaltechPedestrians/,but I can't reproduce the same results reported in this paper " Towards Reaching Human Performance in Pedestrian Detection" .
How we evaluate models on Caltech test using the new annotations?
The Waveform Model of Vowel Perception and Production (WMV, 2009) has achieved human performance multiple times. Below is a short list of achievements so you can suggest next steps.
1) WMV explains vowel perception, production, and perceptual errors (2009).
2) Presented human performance on the Hillenbrand et al. (1995) dataset at an Acoustical Society of America conference (ASA, 2011).
3) Presented human performance on the Peterson and Barney (1952) dataset at an ASA conference in 2014.
4) Achieved human performance on streaming speech (currently being prepared for publication).
5) Recent work has been focused on identifying concussions from h-vowel-d productions (preliminary results presented at an ASA conference in 2014).
As the only working model of cognition that has been successfully introduced into algorithms and achieved human performance on the most cited datasets in the literature, what additional points of validation should be next?
In most of AI research the goal is to to achieve higher than human performance on a single objective,
I believe that in many cases we oversimplify the complexity of human objectives, and therefore I think we should maybe step off improving human performance.
but rather focus on understanding human objectives first by observing humans in the form of imitation learning while still exploring.
In the attachment I added description of the approach I believe could enforce more human like behavior.
However I would like advice on how I could formulate a simple imitation learning environment to show a prove of concept.
One idea of mine was to build a gridworld simulating a traffic light scenario, while the agent is only rewarded for crossing the street, we still want it to respect the traffic rules.
Jasper Busschers master student AI
There are many research in the literature about virtual reality in wayfinding research. Some of them reported that HMDs increase the presence but affect spatial navigation task negatively compared to traditional desktop system. Others reported that HMDs increase presence and provides better spatial navigation performance. However, the technology they used are very poor in terms of resolution, field of view etc. I could not find any research conducted a wayfinding experiment by Oculus Rift. I'm curious about that using Oculus Rift in wayfinding research gives coherent results with real environments or not?
I plan to design a research with pre-test post-test design with study performance as the dependent variable.
Motivation and The Candle Problem
by Rhoads on August 13, 2013 in Motivation
Our world continues to move faster and is increasing in complexity. So what’s the best way to motivate people to think creatively and solve complex problems in a shorter amount of time? Let’s take a look at a study that examines this question.
“The Candle Problem” was conducted by Professor Duncker in 1945. Individuals are led into a room that has a table sitting against the wall. On it is a box of thumbtacks, some matches, and a candle. They are asked to attach a lit candle to the wall in a way that the wax won’t drip onto the table and to do this as fast as they can.
Many who attempted the test tried creative methods but were unsuccessful. Some took a tack and tried to pin the candle to the wall…but that wouldn’t work.
Others lit a match and melted the wax on the side of the candle and then tried to stick it to the wall …but, that wouldn’t work either.
Some, figured out that the real solution was to empty the box of tacks then tack it to the wall and place the lit candle in it.
Professor Duncker realized that people have difficulty in solving a problem when one element has a fixed function that must be altered. In this case, the problem solver has to be able to realize that the box is not just a container for the tacks, but that it can also be used to hold the candle on the wall.
To demonstrate this point, he repeated The Candle Problem again with new participants, but this time he placed the tacks next to the empty box. This time the problem solvers could see that the box’s function was not directly tied to holding the thumbtacks. As a result, virtually all the participants were able to solve the problem with relative ease.
The Candle Problem becomes even more intriguing when you add a monetary reward to the task which is exactly what a professor of psychology, Sam Glucksberg did in 1962.
Glucksberg took a set of new participants and split them into two groups. He told the first group that “the person to complete the problem the fastest will receive $150, and if you are in the top 25% of fastest times you’ll receive $40 (values adjusted for 2013 value).” To the other group he did not offer a monetary reward, but still asked them to solve the problem as fast as they could.
Then he split those groups up and half of each group faced the problem with the tacks placed outside of the box, and the other half faced the problem in the more complex way with the tacks placed in the box.
You’d expect that those who were given the incentive of money would solve the problem faster and in fact that is true when the tacks were out of the box. More of the participants in the group that was incentivized by money solved the problem and with faster times than the other group.
However this was not the case when the problem was given in the more complex way with the tacks placed inside of the box. In this case, those who were not incented by money performed better than those who were offered money.
Glucksberg found that adding the variable of competition for a monetary reward creates levels of stress that shuts down the creative thinking and problem solving areas of the brain.
This finding can have huge implications on the reward model that many organizations use today. Many organizations use an “if – then” motivation model, meaning if you do this, then you get that. While directly tying financial incentives in an “if – then” format to simple tasks can be effective, tying financial incentives in this format to complex tasks that require creative thinking will actually have a negative impact on performance.
This viewpoint was popularized by Daniel Pink in his book Drive where he unpacks the concept of intrinsic and extrinsic motivation. The main premise of his book is that you when you hire people to do complex and creative work, you need to motivate them with autonomy, mastery, and purpose rather than the carrot of a financial reward. Pink is clear to point out that money is still very important…however for this type of work, organizations should pay their employees very well and not tie it directly to the result of their creative work because that would only decrease their performance.
So if you lead a team or are trying to figure out how to motivate people to do something, think about the type of task that is being completed and if it complex and requires creative thinking, understand that using money as a reward for performance on this type of work will actually decrease the quality of the work you are trying to incite.
Drive, by Daniel Pink
The Cornell Daily Sun, “Solve this Problem, Receive $20”, by Steven Zhang
The Candle Problem – Wikipedia
Experts Predict When Artificial Intelligence Will Exceed Human Performance
Dear Researchers and Professors,
Trust you are good.
I'm a researcher at School of Computing at Dublin Institute of Technology and currently working on Human Mental Workload under Dr. Luca Longo (www.luca.longo.eu). My research includes the use of Machine Learning in conjunction to the NASA-TLX and its features. For this purpose, I have started to contact all the scholars who have done research with the NASA-TLX and collected primary data. My goal is to collect as many datasets as possible from different researchers to build a data-driven model that considers a multitude of human tasks. I was wondering whether you can share with me your data. I am aware of new GDPR regulation for data protection and privacy and to stick with it, I am not interested in the names of the participants of your dataset, but only in their answers.
It would be of great help if you could provide us the following: 1. Dataset (anonymous containing only the NASA-TLX answers) 2. Which task/s humans performed during the experiment? (perhaps you have already published a paper which we will cite) 3. What was the task performance measure/s used (time/number of errors/etc..)?
I thank you in advance for your kind availability and I am sorry if I have caused any inconvenience. I hope you can help us with this challenging research activity. I look forward to hearing from you.
Raunak Renge and Dr. Luca Longo School of Computing College of Science and Health Dublin Institute of Technology
I was thinking about what would be truly impossible problems in machine learning, even with unlimited data. I quickly taught of the well known halting problem, which is known to be impossible to decide.
However humans are in many cases able to decide the problem by analyzing code. While my initial instincts are that machine learning could never fully solve the problem, would it be reasonable to believe that it would be possible to improve on human performance?
After all any program can be converted to a turing machine which is a grammar. A sort of compiler can be used to build a meaning representation, which could then be used to train a machine learning application.
And could working on such problems allow us to study better models that incorporate some sort of reasoning capability?
I understand that external factors like smell and colour of a room can affect the mood or performance of a person. But to what level is the effect ? Is there any quantitative measure of its influence.
Has anyone did a real life testing ?
The curves of learning examine the quantitative changes in human performance during the acquisition of a certain skill can be expressed in the improvement of performance as a result of learning processes (through methods and methods of learning), when the learner attempts to access many of the correct learning of skill, these attempts can be expressed in A learning curve is a functional relationship between an independent variable (the type of practice and its amount, usually represented by the number of attempts) and another variable is the performance, the type of performance to be learned. Most learning curves show changes in the rate of improvement.
The performance qualifier is described as what an individual does in his or her
current environment. Since the current environment always includes the overall
societal context, performance can also be understood as "involvement in a life
situation" or "the lived experience" of people in their actual context.
The capacity qualifier describes an individual’s ability to execute a task or an
action. This construct indicates the highest probable level of functioning of a
person in a given domain at a given moment, and to assess the capacity it is said that one would need to have a “standardized environment". But again, it has been mentioned that the capacity qualifier assumes a 'naked person' assessment, that is, the person's capacity without personal assistance or the use of assistive devices.
So what is measured in actual life situations, capacity or performance?
Eg: A patient walks 10 meters in 5 minutes using an AFO and walks 10 meters in 15 minutes without an AFO. What in this case is capacity and performance?
I want to explore and measure terrorism impact on employees performance. What theories are related to terrorism and stress?
We are looking for a way to assess "muscle damage" without an invasive method. Since muscle damaging protocols lead to increased biomarkers (e.g. CK, CRP,...), we need a method to proof that the impact of our exercise protocol induces a real local muscle damage, without taking blood samples.
Is there any scale to measure the performance of employees as well as the performance of the management in the organization. Is there any way or mean to measure the performance? performance and the efficiency are the same thing? any relation or co relation among performance, efficiency and effectiveness? honesty is performance or just ethics ? more amount of work responsibilities lead to more number of failures and less work burden having less chances of failures , so performance of person with more responsibilities is measured on which parameters?
How annual confidential reports (ACRs) of employees in public sectors could be replaced with performance appraisal system or any other innovative system that can replace ACR system?how and why its more appropriate / relevant?
The top management in any organization are the key persons and they are the leaders .But public sector's top management is not gauged and their performance is not measured , no policy exits to monitor or control the performance of top management....its so strange as top management are the real leaders but there is no control over them....
I want to get all of you to please contribute with personal experience in this performance management system innovations and reforms ....
The HRA is an approach widely used in the past for the assessment of the human factor in the safety of the nuclear plants. In your opinion, opportunely changed the models to evaluate it, is it possible to focus this method on the assessment of the stress in the production environment, using for example methods such as the system dynamics or others?
I have difficulties on finding research articles that tells about the relationship or impact of managerial/management expertise on org. performance. I hope someone can help me on this.
Thanks in advance and God Bless..
Can you suggest good scales and questionnaires that measure academic performance or success with university students?
And is there also a corresponding scale that measures performance or success in a workplace?
What equipments do you suggest for the motor behavior or human performance LAB? (for psychological, motor control and learning and biomechanical factors).
Is there any other equipment(s) more valid than the Vienna test systems?
Would you separate your suggestion with each group?
I am working on flying and fatigue management of the crew members, in order to improve our fatigue countermeasures.
I’m looking for different military and civilian aviation regulations that are used in several countries to compare with our French regulations, and different tools that can be used to evaluate the fatigue of pilots and other crew members, before a flight in order to know if it's safe to let them fly, and after the flight. These tools must be quite easy and not too long, in order to have a good involvement of the crew members.