In the past, computer games have been blamed for everything from teenage obesity to acts of murder, and the complaints show no sign of letting up.

Any game which promotes a particular vice, we are often told, predisposes its more impressionable players to commit that vice in their everyday lives. And as each new game gets more immersive and lifelike, so the dividing line between computer fantasy and reality becomes more blurred.

Aside from all the Mary Whitehouse-esque nay-saying, this leads to some rather exotic court cases; such as the time a 45-year-old Belfast man explained robbing a lingerie shop by saying he believed he was a magical elf named Beho. Of course, Beho was the man's in-game alter ego.

Recently, two independent research bodies have cooked-up a new accusation to level at games developers - namely that playing racing games makes you a bad driver.

The first, carried out by driving school BSM, was part of a poll of 1,000 drivers. Each was asked if they took more chances on the road after a gaming session - and 27 percent of those aged under 24 admitted that they did.

The second study was carried out by researchers at Ludwig-Maximilians University in Munich, and took a more exhaustive, three-pronged approach.

The first part of the Munich study echoed the earlier BSM one. 290 people were asked about their driving habits, their accident records and how often they played driving games. The results appeared to link regular simulated racing with a more aggressive, competitive driving style on real roads, and a greater number of RTAs.

The second part involved their watching 15 video sequences of a driver approaching different risky situations, much like the hazard perception section of the UK driving test. Participants were asked to press a key at any point when the risk became too great to ignore.

Of the 68 who took part, those who had not played racing games beforehand performed better when it came to spotting the risks and pressing the key.

For the final section of the study, 83 people were told to play either racing or 'neutral' (i.e. football) games for a set period of time and then asked how they were feeling. Typical responses from the racing set indicated a desire to go out and take risks, whereas the neutral gamers were left unaffected.

So there you have it: four separate results from two independent research groups, all of which suggest that playing a little Gran Turismo is going to turn you into a boy racer.

It's just a shame there are quite so many obvious flaws in the theory.

For starters, what the BSM study fails to take into account is that young male drivers in their teens and early twenties are practically predisposed to taking risks, which is why young driver car insurance premiums for this age group are so ridiculously high. Sure they're impressionable and the sight of fast cars in a computer game might spur them on; but so would any film with a car chase or even the sight of another driver speeding by.

Computer games taken on their own represent at worst a small fraction of the reason young drivers take risks at the wheel. Singling out their influence is thus both short-sighted and rather unhelpful.

The root of the problem with the Munich experiments, on the other hand, is the fact of their being so hopelessly dependant on the technology they set out to discredit.

For instance: why would one expect somebody who has been playing a computer racing game to sit down at a different computer - running a much less immersive driving simulation - and take it as seriously as real driving?

Driving and pressing a key at the right moment are two very different things - and so the claim that simulated racing affects your performance in a real car remains unfounded on the basis of the second test.

As this was the only one of the three to directly test so-called 'driving' abilities - the remaining sections dealing instead with memories and emotional responses - the results of the Munich study amount to little more than a suspicion about racing games and driving performance.

They've perhaps proved that an activity which features simulated risk-taking makes us think about taking risks - but surely we could all have guessed that much beforehand?

In fact, today's drivers have got much more to gain from computers than they have reason to fear them. Besides carrying out a lot of the research which keeps us safe on the roads, through improved vehicle designs and greater public awareness of the dangers, computers also help mechanics diagnose car trouble and make our driving theory tests more thorough.

And of course, without computers we'd all still be ringing round twenty or thirty different car insurance firms every time we wanted a competitive car insurance quote.