The LIDA architecture was primarily developed by Stan Franklin and colleagues at the University of Memphis and is scientifically founded in cognitive science and neuroscience. In addition to supporting control frameworks for software agents and robotics, the architecture can provide hypotheses to direct future studies. In addition to providing plausible explanations for numerous cognitive processes, the LIDA conceptual model is meant as a tool for contemplating how minds operate.

Cognitive cycle

We can break down the LIDA cognitive cycle into three parts: 

• the understanding phase, 
• the attention (consciousness) phase, and 
• the action selection and learning phase. 

At the start of the understanding phase, low-level feature detectors in sensory memory are turned on by incoming stimuli. The output activates perceptual associative memory, where higher-level feature detectors feed into more abstract entities like objects, categories, actions, events, etc. The resulting percept moves to the Workspace, which triggers both Transient Episodic Memory and Declarative Memory to make local associations. When these local connections and the percept are put together, they make up the agent's current situational model, which is what the agent thinks is going on right now. The attention phase starts with the essential parts of the current situational model coming together to form coalitions. 

These coalitions compete for attention, which means a spot in the current conscious content. Then, these conscious contents are sent worldwide, which starts the learning and action selection phase. As the conscious broadcast reaches perceptual, episodic, and procedural memory, it creates new entities and associations and strengthens old ones. While all this learning is going on, possible action plans are pulled from Procedural Memory and sent to Action Selection, where they compete to be chosen as the behaviour for this cognitive cycle. Finally, the cognitive cycle is finished when the chosen behaviour tells the sensory-motor memory to devise an algorithm for doing it.

Conceptual model

Two hypotheses support the LIDA architecture and its accompanying conceptual model: 

1) A significant portion of human cognition is facilitated by cognitive cycles, interactions that frequently occur (10 Hz) between conscious contents, multiple memory systems, and action choices. 

2) These cognitive cycles are the "atoms" of cognition that comprise higher-level cognitive processes.

Conclusion

LIDA is a hybrid architecture because it uses multiple computing techniques. Modules and processes using these pathways make up the LIDA cognitive cycle.

In addition, the LIDA architecture seeks to mimic a sizable piece of human cognition by providing a complete, conceptual and computational framework. The LIDA architecture is an attempt to implement and elaborate on several psychological and neuropsychological theories, such as 

• the Global Workspace Theory, 
• situated cognition, 
• perceptual symbol systems, 
• working memory, 
• memory by affordances, 
• long-term working memory, and 
• the H-CogAff architecture, 

It encompasses a wide range of cognitive modules and processes.

Furthermore, incorporating computational principles from the "new AI," the LIDA architecture consists of several components. Some examples of this are the schema mechanism, the Behavior Net, and the subsumption and Copycat architectures.