The disconnect between knowing what and how

The most noticeable distinction between procedural and declarative knowledge/memory is the fact that procedural knowledge is usually stored implicitly and semantic knowledge is usually stored explicitly. This means we can access procedural knowledge without conscious effort but the same is not usually true for declarative knowledge.

This can cause interesting effect when procedural and declarative memory don't match up meaning people can remember how to do something but do not remember what they did. For example, asking a highly skilled jazz musician to explain what they did in their complex improvised solo. #

Therefore it is useful for us to explore situations in which procedural and declarative are related/ unrelated to understand how useful this distinction truly is.

(Pammer and Blink, 2018)

Aim of the study: to measure visuo-cognitive mechanisms from expert drivers from competent drivers.

• Did this by comparing expert drivers e.g. ambulance drivers to competent drivers who drive day to day.

Method:

Non-driving visuals and cognitive tasks that will underlie good driving skills such as:

• scanning environments for targets

• tracking objects

• identifying unexpected objects

• tendency towards intrusive thought patterns

Minimising Factors:

• top down factors such as familiarity to driving, or scenery

Conclusion:

• Expert drivers performed better

• Will be due to superior skills for core visuo-cognitive constructs.

How This Links:

• This links with the knowing how vs. knowing what, where due to ambulance workers needing to frequently drive with high speeds and through traffic to get to their destination they will know how to complete this safely.

• Whereas, competent drivers will know what to do to drive fast and if traffic was letting them through they will know what to do to cut this traffic.

• However, the difference between these and expert drivers is that the expert drivers know how to do this safely whereas competent drivers even with knowing what to do, majority of these will not know how to do perform this without any negative consequences.

• This is shown as the difference with crash rates between competent and expert drivers using these visual processing expertise is significantly different.

• With a total of 15,358 crashes only these involving 10 (less than 1%)

  from emergency service vehicles even with these expert drivers having to undergo more dangerous driving experiences.

Conclusions

• This study only measured cognitive-perceptual mechanisms using artificial testing techniques rather than real driving experiences of these drivers with secondary sources of car crash rates.

• Whereas, more studies are needed to explore real driving experiences without using artificial methods.

• However, this does not mean this study does not give valuable insight on the topic of 'knowing how vs knowing what'. As it does show the differences between the two.

References

Pammer, K., & Blink, C. (2018). Visual processing in expert drivers: What makes expert drivers expert? Transportation Research Part F: Traffic Psychology and Behaviour, 55, 353–364. https://doi.org/10.1016/j.trf.2018.03.009

‌

Moravec’s Paradox originates from the fascination with artificial intelligence and the attempt to replicate the complex abilities of humans in non-human formats. As researchers worked to create robots, they discovered an intriguing paradox: tasks that humans find relatively simple - like walking, brushing teeth, or expressing emotions - are incredibly difficult for robots to perform. On the other hand, tasks that humans might struggle with, such as playing chess or solving mathematical equations, are areas where artificial intelligence excels.

This paradox can be explained through Gilbert Ryle's distinction between declarative and procedural knowledge, which provides insight into the different types of knowledge and skills required for various tasks. Let’s break this down in relation to Moravec’s Paradox: 

Declarative Knowledge (Knowing What): 

Robots excel at tasks like playing chess or solving math puzzles because these activities rely on declarative knowledge—facts, rules, and patterns that can be explicitly programmed. Humans, however, often find these tasks challenging without consistent practice, as they require higher-level cognitive reasoning based on clear, structured information. 

Procedural Knowledge (Knowing How): 

In contrast, humans excel at tasks involving procedural knowledge, such as perception and motor skills—e.g., picking up a coin. These tasks rely on unconscious, automatic abilities that are hard to explain or codify into algorithms. Robots struggle with such tasks because they require real-time sensory processing and adaptive actions, which are not easily captured by mathematical models. 

Benefits of these distinctions  

The benefits of distinguishing between these two types of knowledge, declarative knowledge and procedural knowledge, are clear. By recognising and separating these skills, we can target them individually to optimize performance and achieve better outcomes. This distinction has important implications for both human development and the study of robotics. For example, research suggests that the more closely robots are programmed to mimic human genetic and cognitive structures, the better their performance in complex tasks. Similarly, when we focus on enhancing a human’s declarative knowledge, we see improvements in abilities such as mathematical reasoning and problem-solving 

The aim of this study was to measure visuospatial- cognitive mechanisms of expert drivers and compare them to experienced and competent drivers, gaining an understanding of how expert drivers may differ from non expert drivers.

Non-driving visuals were used alongside cognitive tasks that should identify good driving skills e.g. scanning the environment for targets and tracking multiple objects, identifying unexpected objects. This design is aimed to minimise the influence of top-down factors such as familiarity allowing there to be a focus on influences that are stimulus driven.

Expert drivers performed better than non-expert drivers when completing tasks designed to reflect on task performance however, in visual search and noticing unexpected objects to then track these objects, expert drivers were the same as the two younger driver groups.

What do these results suggest?

When it comes to driving- expertise is qualitatively different from experience, and driving expertise may be partially derived from superior skill in underlying core visuo- cognitive constructs.

Application: This finding is importantly for understanding driver training programmes allowing there to be implication be made into a safer world of drivers. It identifies the underling cognitive-perceptual networks within driving and certain demographics.

- However, this perspective has been critiqued, primarily from an intellectualist perspective, from various aspects.

1.      Intellectualists argue that all forms of knowledge are inherently intellectual.

• Knowing THAT precedes knowing HOW; practical knowledge is underpinned by a strong intellectual understanding.

• In a medical context, performing an appendectomy requires a deep understanding of bodily anatomy, surgical instruments, and techniques. This guides the knowing HOW, through a comprehensive knowledge of context-specific facts.

• Therefore, Ryle’s view that knowing HOW does not involve intellectual content can be seen as an oversimplification; real world experiences demonstrate the need for practical knowledge that is underpinned by intellectual knowledge.

• Links to the Expert Drivers study; this was done in a controlled environment, and doesn’t show how the two types of knowledge are married in real-world scenarios. Driving requires knowing road rules, vehicle mechanics, and responses to hypothetical scenarios; an intellectualist might argue that this forms the foundation of skilled practice, once again highlighting that knowing how is inextricably tied to knowing what.

-         SOURCE; Shallice, T (1998) ‘From Neuropsychology to Mental Structure.’ Shallice’s exploration of cognitive structures, arguing that that expertise is grounded in cognitive networks and structures. This shows the depth of the relationship between knowing what and knowing how at a cognitive level.

• Ryle’s theory also undermines the intellectual underpinnings of so-called ‘automatic knowledge.’ Whilst some skills might feel mindless upon execution, an intellectualist would argue that it stems from underlying intellectual processes (perception, anticipation, judgement).