Scientists Identify Transition Between Easy and Difficult Tasks

January 12, 2010 By Lisa Zyga feature
Scientists have found that the nervous system uses different control mechanisms for easy and difficult aiming tasks, in order to optimize motor performance. Hand drawn by Hans Holbein the Younger around year 1523.

( -- One of the primary ways in which we as humans can manipulate our environment involves working with our hands, using them to point, reach, and handle tools such as pens, needles, and screwdrivers. While it's clear from everyday observation that it takes longer to perform difficult aiming tasks (those that require precise, detailed movements) than easy ones, scientists have long thought that performance time increases linearly with task difficulty. However, a new study shows that the human nervous system is more complicated than this theory suggests.

In their study, Raoul Huys, Laure Fernandez, Reinoud Bootsma, and Viktor Jirsa from the University of the Mediterranean in Marseille, France, have investigated the relationship between task difficulty and movement speed. Until now, this relationship has been thought to be continuous, as expressed by Fitts' law. Proposed in 1954 by Paul Fitts, the law states that the time needed to successfully perform an aiming movement increases linearly with task difficulty.

However, Huys and his coauthors found something different: that the relationship between task difficulty and movement speed is discontinuous. The researchers identified a specific level of task difficulty at which a transition occurs, where the human nervous system abruptly engages a different for difficult tasks. Specifically, humans seem to use faster rhythmic movements when performing easy tasks, and slower discrete movements when performing difficult tasks.

“Of greatest significance is the finding that two control mechanisms are involved in reciprocal precision aiming and that each of them adheres to its own particular speed-accuracy trade-off,” Huys told “This puts the well-known speed-accuracy trade-off in a completely different light than previously held. It also tells us that evolution has endowed us with different functional modules that each adhere to its own merits and limitations, and that we can use them in order to optimize motor performance given certain task constraints.”

The scientists arrived at these conclusions after performing experiments in which participants moved a cursor back and forth between two targets on a screen as quickly and accurately as possible. There were 12 different versions of the test, with 12 different target widths to represent varying levels of task difficulty. The researchers measured each participant’s total movement time, as well as its constituents, acceleration time and deceleration time, which were defined as the duration prior to and following peak velocity, respectively. The scientists found that total time and deceleration time always increased as task difficulty increased. However, acceleration time stopped increasing at a certain degree of difficulty, revealing a discontinuity between easy and difficult tasks.

This transition seems to mark the point at which a different control mechanism becomes engaged, evoking a change from rhythmic to discrete movements. The results suggest that, when moving rhythmically, both acceleration time and deceleration time increas with increasing task difficulty. But after the transition to discrete movements, only the deceleration time increased. Thus, acceleration time is not a continuous function of the degree of difficulty, implying that Fitts’ law is discontinuous due to a change in the control mechanism. The researchers think that this discontinuity has not been observed before because most experiments have focused primarily on the more difficult discrete movement tasks while overlooking easier rhythmic tasks. As Huys explained, the discovery of the two control mechanisms could have applications in man-machine interfaces, robotics, and sensorimotor rehabilitation.

“In robotics, the two control mechanisms have sometimes been implemented,” Huys said. “Our result indicates that which mechanism should be indicated depends on the speed and accuracy requirements of the task to be fulfilled. In addition, clinicians are still seeking for easy to be implemented clinical assessment tools in the context of sensorimotor rehabilitation and neurological and motor pathology evaluation. We hope that our result may help their development.”

In the future, the scientists hope to further explore how neurological and motor pathologies affect the utilization of the control mechanisms.

“The first population that we will investigate is patients suffering from Parkinson’s disease,” Huys said. “Thereto, one of my coauthors, Laure Fernandez, has recently initiated a collaboration with the La Timone Hospital here in Marseille. Another future avenue that we would like to pursue is to investigate how the of (healthy) humans implements the two motor control mechanisms.”

More information: Raoul Huys, Laure Fernandez, Reinoud J. Bootsma, and Viktor K. Jirsa. “Fitts’ law is not continuous in reciprocal aiming.” Proceedings of the Royal Society B. doi:10.1098/rspb.2009.1954


Related Stories

Recommended for you

A new compound targets energy generation, thereby killing metastatic cells

October 17, 2017
Cancer can most often be successfully treated when confined to one organ. But a greater challenge lies in treating cancer that has metastasized, or spread, from the primary tumor throughout the patient's body. Although immunotherapy ...

New approach helps rodents with spinal cord injury breathe on their own

October 17, 2017
One of the most severe consequences of spinal cord injury in the neck is losing the ability to control the diaphragm and breathe on one's own. Now, investigators show for the first time in laboratory models that two different ...

Pair of discoveries illuminate new paths to flu and anthrax treatments

October 17, 2017
Two recent studies led by biologists at the University of California San Diego have set the research groundwork for new avenues to treat influenza and anthrax poisoning.

New method to measure how drugs interact

October 17, 2017
Cancer, HIV and tuberculosis are among the many serious diseases that are frequently treated with combinations of three or more drugs, over months or even years. Developing the most effective therapies for such diseases requires ...

Research finds that zinc binding is vital for regulating pH levels in the brain

October 17, 2017
Researchers in Oslo, Norway, have discovered that zinc binding plays an important role in the sensing and regulation of pH in the human brain. The findings come as one of the first studies that directly link zinc binding ...

Researchers find factor that delays wound healing

October 17, 2017
New research carried out at The University of Manchester has identified a bacterium—normally present on the skin that causes poor wound healing in certain conditions.


Adjust slider to filter visible comments by rank

Display comments: newest first

2.3 / 5 (6) Jan 12, 2010
WOW, I can't believe that these guys are spending precious resources studying this. Seems pretty obvious to me that the discontinuity occurs at the point in time where the individual being tested loses faith in their ability to react and goes into thinking mode instead of reacting mode. How is this new ground? Anyone? Anyone at all?

And, as an aside, why is it that almost every one of these articles starts with the same number of votes as the number of people cited in the article and that all of these votes always rate the article a 5?

Coincidence? I doubt it............LOL
3 / 5 (4) Jan 12, 2010
It's a first step towards finding out where the higher brain functions set in to take control of any task with relation to difficulty.

Just think of how much more efficent one could structure any learning task (e.g. schoolwork) if we were to know exactly how hard the tasks should be.

Stuff that kids learn 'on automatic' goes little towards increasing their mental abilities. Stuff that is too hard is frustrating. If we could keep task difficulties at just that level slightly above the discontinuity then learning experiences would be far more rewarding and effective.
5 / 5 (1) Jan 12, 2010
Well -- what about the point at which you conscienously spend time to think about your intended action.

With trivial actions you may think about them the first couple times but after that you delegate the action to purely muscle and to increase performance you try to treat it as a reflex action. This is like tying your shoes and why it is actually hard to explain to children - because you stopped thinking about how to do it a long time ago, it is now been deletaged to trivial and a muscle memory. Or tying a tie. -- but most people still think about tying a bow tie-- or delibritely creating a slip not-- or hitting snooze on an alarm clock.

Harder tasks demand you pause -- stop accelerating -- and think about the action to perform it corectly. -- adding two digits together is easy for most adults, but what about being given five digits that are randomly colored and only adding together the ones that are green -- that takes thought.

I think they should focus on the pause.
2.3 / 5 (3) Jan 12, 2010
antialias---like most of your comments, I find this one an intriging mixture of high intelligence and no common sense. Your theory of action is great in concept until you actually try to impose it on the system which is not made up of your average person but is made up of individuals whose levels of "hard" vary so much as to make this an impossible task. Ask any teacher that you know.

El Nose--exactly my point, thank you for clarifying
3.3 / 5 (3) Jan 12, 2010
jonnyboy: I think then you aren't reading what I write. I didn't say anywhere that this should be geared towards creating a system that should be used equally for all (e.g. like a standardized rule - which would be stupid).

It could be that this way we can set up tests that will allow us to judge where the individual's point of 'deliberate thought' begins. Think about how a lot of schooling in the future may be online with task difficulty being adapted to the level of the student on the fly.

If we can simply track how long it takes for a student to complete a task as opposed to their average time taken this might give an indication on how to scale the difficulty of the next question. This is just one off-the-top-of-my head uses - but it alone would revolutionize education.
How often have you been bored out of your skull in classes/lectures? How often did the lecture go over your head and were just wasting your time? One could cut that waste and focus on each student's 'sweet spot'.
not rated yet Jan 13, 2010
WOW, I can't believe that these guys are spending precious resources studying this. Seems pretty obvious to me that the discontinuity occurs at the point in time where the individual being tested loses faith in their ability to react and goes into thinking mode instead of reacting mode. How is this new ground? Anyone? Anyone at all?

And, as an aside, why is it that almost every one of these articles starts with the same number of votes as the number of people cited in the article and that all of these votes always rate the article a 5?

Coincidence? I doubt it............LOL

Well in my work trying to compose piano music, I find I always seem to compose music just out of my reach technically (no Horowitz for sure) and there is a definite discontinuity in my search for better fingering, more accurate placement, etc. I always wonder why it is so hard to get recordings right, play something through all the way without mistakes, Practice, practice, practice I guess.
1 / 5 (1) Jan 13, 2010
they are showing the point where talent and comparitive advantage of the individual resides. given that this (like so many other empirical facts) tends to grate against the marxian collective, it will be interesting to see how they interpret things in the near future.
not rated yet Jan 17, 2010
"... if we were to know exactly how hard the tasks should be."

Should be? Odds are this differs from person to person, as does learning style. Teaching a person to recognize that transition point, now there's the rub.
not rated yet Jan 19, 2010
boxfiddler you took the words right out of my mouth. I would wager, odds would have little to do with it differing from person to person, but I know it will.

If you've ever taught a class of 20+ students, you'll find 3 or 4 people that just need it framed in a certain way to be understood and 2 or 3 people bored because it's slow.

There is no correct speed of learning for a group and I hardly think one could quantify it in humans.

It will differ from subject to subject. Task to task. Cognitive or physical.

This seems more useful in building neural networks in robots to more closely mimic a humans actions when performing dexterous tasks...

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.