This post provides a shorter, less technical introduction to this material. For the full version with references, see part 1. I also focus here just on the bike, but the lessons here apply to any activity involving skilled movements: dance, rock climbing, baseball, high jump, etc.
New riders at the velodrome, watching film, are often surprised at how high they sit on the bike. They feel as if their head is more tucked towards their handlebars than it actually is. Similarly, those who ride based on perceived effort can be in for a shock once they get a powermeter. What feels to them like a nine-out-of-ten might only be 70% of their VO2max power.
This lack of bodily awareness hurts performance on the bike. Sitting high means more air resistance and less speed for the same energy spent. Pushing 70% VO2max when your program calls for 90% means a suboptimal training effect. Examples are easily multiplied. Consider hill climbing, which requires both proper pacing of effort and coordinated pedalling out of the saddle. Get these factors wrong and you’ll find yourself walking to the top — something we all want to avoid, even on a cafe ride with friends. In my discipline of track racing, bodily awareness is especially important. Track involves specialized skills like sprinting out of the saddle down the steep bank of the velodrome, or accelerating the bike as quickly as possible out of a start gate. These take highly coordinated body movements.
Why is bodily awareness so poor, and how can one improve it? First we need to separate the issue of poor bodily awareness from the question of proper technique. For example, most riders don’t feel that their legs fail to relax fully on the pedal upstroke, thereby leading to a loss of power. Force from the downstroke not only works to propel the bike forward, but also must overcome resistance from the back leg. But, as studies show, technique adjustments to relax on the upstroke (like pulling up on the pedals) actually lead to less efficient pedaling. Similarly, good bike fitters know that getting someone as low as possible on their bike may not be optimal, and there are differing schools of thought on how to climb (e.g., in or out of the saddle).
Whatever the optimal technique in these and other cases, the problem is that most riders aren’t aware of how their own movements deviate from this ideal. For instance, they don’t even feel that their legs resist the pedals on the upstroke, or that they’re sitting tall. If you don’t accurately perceive your own body position, movement, or effort, then you won’t detect and correct flaws in your technique.
Let’s start with what’s most familiar to riders: their own subjective experience of their body. Just as our external senses like vision bring the outside world into our stream of consciousness, so too our internal senses like proprioception bring our own bodies into our stream of consciousness. Right now, as I sit here typing, I know without looking that my knees are bent at roughly a 90-degree angle. I know this because I experience my body through proprioception.
There are a few key features of our proprioceptive bodily experience which explain why bodily awareness is often so poor. Consider what it’s like to pedal your bike. If you pay attention the next time you’re out riding, you’ll probably notice that within your experience you can focus on three things: (1) “sensations” like twinges of pain or the burn of lactic acid, (2) the position of your body, and (3) the bike itself. What’s curious is that your sensations largely do not correlate with your experience of your body position or the bike. For example, as I pedal, I know when each foot is at the bottom of the stroke, but there’s no sensation which flashes in my stream of consciousness telling me my foot is at the bottom. Likewise, I can feel that my saddle is a hard piece of plastic (as opposed to being well cushioned), but again there are no distinct sensations through which I feel the saddle’s hardness. I just experience my leg position and the saddle directly.
Work from neuroscience helps explain why we experience our bodies directly, without experiencing them through qualitative sensations. The brain keeps track of the body through a number of different representations, or models, in the cortex and cerebellum. Sensory receptors in your muscles, tendons, and ligaments (called proprioceptors) send information to the brain about the configuration of your body, but these signals aren’t what drive the various body models in the brain. Instead, the brain’s body models are built around copies of motor commands. In effect, the brain sends out commands telling your muscles to contract, then predicts the movements which should result from those commands. The brain merely uses proprioceptive feedback to fine-tune and correct the body models it has constructed from its own motor commands.
So why is bodily awareness often poor? The answer isn’t simple, but what’s been said so far hints at the basics. First, we have few if any conscious qualitative sensations to clue us into our body position. Riders looking to discern their position based on how they “feel” are likely to find either no relevant feeling at all, or misinterpret what few sensations are there. Second, bodily experience depends on the brain’s construction of body models, models which it builds by predicting the consequences of motor commands. A rider’s brain might just be bad at making these predictions, either because the movements are new, or because it has never received the sort of feedback required to correct them. If so, the brain will produce inaccurate body models.
Thankfully, bodily awareness is something that can be improved. It’s not quite true that there are never sensations which correlate with bodily experience, and this fact gives us the first hint at how to improve bodily awareness. Try, as an experiment, pedalling with your knees splayed slightly to the outside (assuming you have no knee conditions this might aggravate). You’re likely to notice that this new movement pattern brings out new sensations which correlate with the new movement. Where before it was hard to attend to any sensations indicating your body position, you now have elicited sensations signalling your outwardly splayed knees.
While this particular example hasn’t been studied scientifically, given what we know from other work a plausible explanation is that this new movement brings about new sensations because it is unexpected. When the brain’s body models successfully predict the incoming proprioceptive signals, those signals are cancelled out. But new movements cause new, unexpected proprioceptive signals that make it to the brain. With these new signals the brain can update and refine its body models.
The upshot is that if you want to improve bodily awareness, you need to help your brain refine and train its body models. This means introducing new feedback and stimulation which the brain can use to do that refining. One way to do this training is by exploring the way changes in your movement patterns generate new sensations. To take a simple example, a rider who experiences their body as being low when they’re really sitting up high should (on a trainer) move their torso through a range of positions: starting fully upright, and pull yourself as close to the handlebars as you can get comfortably. As you move through this range of motion pay attention to how your body feels, feeling for new sensations and paying attention to how those sensations covary with your position. The basic idea is that you want to explore (in a safe manner) how your proprioceptive sensations change with movement, thereby learning to associate the right sensations with each position.
This kind of bodily self exploration can only take you so far. It works only to the extent that you know your body position through some other means, e.g. by seeing your body in a mirror. After all, a simple qualitative sensation itself doesn’t mean anything — it doesn’t tell you anything about your position. At the neural level, the proprioceptive signals generating these sensations are only encoding crude information about things like muscle stretch and joint angle — they don’t encode rich spatial information about your body’s position in objective space. Thankfully, our brains aren’t restricted to just proprioceptive information. They can (and do) refine body models based on any available sensory information. This is why (for example) watching yourself perform movements in a mirror is helpful: the brain uses that visual information to update its body models.
When it comes to bike riding, it’s of course rare that we can use visual cues. Someone racing around the track can’t look in a mirror as they go, nor can someone modulate their hill-climb pedalling on-the-fly by watching video. We need our visual attention focused on the environment while we ride. So it turns out that audio feedback works well — sometimes even better — in place of vision. This could take the form of a coach shouting cues (“lower!”), all the way up to sophisticated sonification devices that convert data from motion sensors into audio tones.
Awareness of your position and movement on the bike is a skill to be learned and improved. Trusting your naive “sensations” won’t be enough. You have to help your brain learn to interpret properly incoming proprioceptive information, and provide yourself with multimodal feedback from visual and auditory cues. Taking time to actively explore how your proprioceptive sensations change with body position can be a good way to start this process of learning.